Shortcuts

mmocr.apis

mmocr.apis.init_detector(config, checkpoint=None, device='cuda:0', cfg_options=None)[源代码]

Initialize a detector from config file.

参数
  • config (str or mmcv.Config) – Config file path or the config object.

  • checkpoint (str, optional) – Checkpoint path. If left as None, the model will not load any weights.

  • cfg_options (dict) – Options to override some settings in the used config.

返回

The constructed detector.

返回类型

nn.Module

mmocr.apis.model_inference(model, imgs, ann=None, batch_mode=False, return_data=False)[源代码]

Inference image(s) with the detector.

参数
  • model (nn.Module) – The loaded detector.

  • imgs (str/ndarray or list[str/ndarray] or tuple[str/ndarray]) – Either image files or loaded images.

  • batch_mode (bool) – If True, use batch mode for inference.

  • ann (dict) – Annotation info for key information extraction.

  • return_data – Return postprocessed data.

返回

Predicted results.

返回类型

result (dict)

mmocr.core

evaluation

mmocr.core.evaluation.compute_f1_score(preds, gts, ignores=[])[源代码]

Compute the F1-score of prediction.

参数
  • preds (Tensor) – The predicted probability NxC map with N and C being the sample number and class number respectively.

  • gts (Tensor) – The ground truth vector of size N.

  • ignores – The index set of classes that are ignored when reporting results. Note: all samples are participated in computing.

mmocr.core.evaluation.eval_hmean(results, img_infos, ann_infos, metrics={'hmean-iou'}, score_thr=0.3, rank_list=None, logger=None, **kwargs)[源代码]

Evaluation in hmean metric.

参数
  • results (list[dict]) – Each dict corresponds to one image, containing the following keys: boundary_result

  • img_infos (list[dict]) – Each dict corresponds to one image, containing the following keys: filename, height, width

  • ann_infos (list[dict]) – Each dict corresponds to one image, containing the following keys: masks, masks_ignore

  • score_thr (float) – Score threshold of prediction map.

  • metrics (set{str}) – Hmean metric set, should be one or all of {‘hmean-iou’, ‘hmean-ic13’}

返回

float]

返回类型

dict[str

mmocr.core.evaluation.eval_hmean_ic13(det_boxes, gt_boxes, gt_ignored_boxes, precision_thr=0.4, recall_thr=0.8, center_dist_thr=1.0, one2one_score=1.0, one2many_score=0.8, many2one_score=1.0)[源代码]

Evaluate hmean of text detection using the icdar2013 standard.

参数
  • det_boxes (list[list[list[float]]]) – List of arrays of shape (n, 2k). Each element is the det_boxes for one img. k>=4.

  • gt_boxes (list[list[list[float]]]) – List of arrays of shape (m, 2k). Each element is the gt_boxes for one img. k>=4.

  • gt_ignored_boxes (list[list[list[float]]]) – List of arrays of (l, 2k). Each element is the ignored gt_boxes for one img. k>=4.

  • precision_thr (float) – Precision threshold of the iou of one (gt_box, det_box) pair.

  • recall_thr (float) – Recall threshold of the iou of one (gt_box, det_box) pair.

  • center_dist_thr (float) – Distance threshold of one (gt_box, det_box) center point pair.

  • one2one_score (float) – Reward when one gt matches one det_box.

  • one2many_score (float) – Reward when one gt matches many det_boxes.

  • many2one_score (float) – Reward when many gts match one det_box.

返回

Tuple of dicts which encodes the hmean for the dataset and all images.

返回类型

hmean (tuple[dict])

mmocr.core.evaluation.eval_hmean_iou(pred_boxes, gt_boxes, gt_ignored_boxes, iou_thr=0.5, precision_thr=0.5)[源代码]

Evaluate hmean of text detection using IOU standard.

参数
  • pred_boxes (list[list[list[float]]]) – Text boxes for an img list. Each box has 2k (>=8) values.

  • gt_boxes (list[list[list[float]]]) – Ground truth text boxes for an img list. Each box has 2k (>=8) values.

  • gt_ignored_boxes (list[list[list[float]]]) – Ignored ground truth text boxes for an img list. Each box has 2k (>=8) values.

  • iou_thr (float) – Iou threshold when one (gt_box, det_box) pair is matched.

  • precision_thr (float) – Precision threshold when one (gt_box, det_box) pair is matched.

返回

Tuple of dicts indicates the hmean for the dataset

and all images.

返回类型

hmean (tuple[dict])

mmocr.core.evaluation.eval_ner_f1(results, gt_infos)[源代码]

Evaluate for ner task.

参数
  • results (list) – Predict results of entities.

  • gt_infos (list[dict]) – Ground-truth information which contains text and label.

返回

precision,recall, f1-score of total

and each catogory.

返回类型

class_info (dict)

mmocr.core.evaluation.eval_ocr_metric(pred_texts, gt_texts)[源代码]

Evaluate the text recognition performance with metric: word accuracy and 1-N.E.D. See https://rrc.cvc.uab.es/?ch=14&com=tasks for details.

参数
  • pred_texts (list[str]) – Text strings of prediction.

  • gt_texts (list[str]) – Text strings of ground truth.

返回

float]): Metric dict for text recognition, include:
  • word_acc: Accuracy in word level.

  • word_acc_ignore_case: Accuracy in word level, ignore letter case.

  • word_acc_ignore_case_symbol: Accuracy in word level, ignore

    letter case and symbol. (default metric for academic evaluation)

  • char_recall: Recall in character level, ignore

    letter case and symbol.

  • char_precision: Precision in character level, ignore

    letter case and symbol.

  • 1-N.E.D: 1 - normalized_edit_distance.

返回类型

eval_res (dict[str

mmocr.utils

class mmocr.utils.Registry(name, build_func=None, parent=None, scope=None)[源代码]

A registry to map strings to classes.

Registered object could be built from registry. .. rubric:: 示例

>>> MODELS = Registry('models')
>>> @MODELS.register_module()
>>> class ResNet:
>>>     pass
>>> resnet = MODELS.build(dict(type='ResNet'))

Please refer to https://mmcv.readthedocs.io/en/latest/understand_mmcv/registry.html for advanced usage.

参数
  • name (str) – Registry name.

  • build_func (func, optional) – Build function to construct instance from Registry, func:build_from_cfg is used if neither parent or build_func is specified. If parent is specified and build_func is not given, build_func will be inherited from parent. Default: None.

  • parent (Registry, optional) – Parent registry. The class registered in children registry could be built from parent. Default: None.

  • scope (str, optional) – The scope of registry. It is the key to search for children registry. If not specified, scope will be the name of the package where class is defined, e.g. mmdet, mmcls, mmseg. Default: None.

get(key)[源代码]

Get the registry record.

参数

key (str) – The class name in string format.

返回

The corresponding class.

返回类型

class

static infer_scope()[源代码]

Infer the scope of registry.

The name of the package where registry is defined will be returned.

示例

# in mmdet/models/backbone/resnet.py >>> MODELS = Registry(‘models’) >>> @MODELS.register_module() >>> class ResNet: >>> pass The scope of ResNet will be mmdet.

返回

The inferred scope name.

返回类型

scope (str)

register_module(name=None, force=False, module=None)[源代码]

Register a module.

A record will be added to self._module_dict, whose key is the class name or the specified name, and value is the class itself. It can be used as a decorator or a normal function.

示例

>>> backbones = Registry('backbone')
>>> @backbones.register_module()
>>> class ResNet:
>>>     pass
>>> backbones = Registry('backbone')
>>> @backbones.register_module(name='mnet')
>>> class MobileNet:
>>>     pass
>>> backbones = Registry('backbone')
>>> class ResNet:
>>>     pass
>>> backbones.register_module(ResNet)
参数
  • name (str | None) – The module name to be registered. If not specified, the class name will be used.

  • force (bool, optional) – Whether to override an existing class with the same name. Default: False.

  • module (type) – Module class to be registered.

static split_scope_key(key)[源代码]

Split scope and key.

The first scope will be split from key.

实际案例

>>> Registry.split_scope_key('mmdet.ResNet')
'mmdet', 'ResNet'
>>> Registry.split_scope_key('ResNet')
None, 'ResNet'
返回

The first scope. key (str): The remaining key.

返回类型

scope (str, None)

class mmocr.utils.StringStrip(strip=True, strip_pos='both', strip_str=None)[源代码]

Removing the leading and/or the trailing characters based on the string argument passed.

参数
  • strip (bool) – Whether remove characters from both left and right of the string. Default: True.

  • strip_pos (str) – Which position for removing, can be one of (‘both’, ‘left’, ‘right’), Default: ‘both’.

  • strip_str (str|None) – A string specifying the set of characters to be removed from the left and right part of the string. If None, all leading and trailing whitespaces are removed from the string. Default: None.

mmocr.utils.build_from_cfg(cfg, registry, default_args=None)[源代码]

Build a module from config dict.

参数
  • cfg (dict) – Config dict. It should at least contain the key “type”.

  • registry (Registry) – The registry to search the type from.

  • default_args (dict, optional) – Default initialization arguments.

返回

The constructed object.

返回类型

object

mmocr.utils.collect_env()[源代码]

Collect the information of the running environments.

mmocr.utils.convert_annotations(image_infos, out_json_name)[源代码]

Convert the annotation into coco style.

参数
  • image_infos (list) – The list of image information dicts

  • out_json_name (str) – The output json filename

返回

The coco style dict

返回类型

out_json(dict)

mmocr.utils.drop_orientation(img_file)[源代码]

Check if the image has orientation information. If yes, ignore it by converting the image format to png, and return new filename, otherwise return the original filename.

参数

img_file (str) – The image path

返回

The converted image filename with proper postfix

mmocr.utils.get_root_logger(log_file=None, log_level=20)[源代码]

Use get_logger method in mmcv to get the root logger.

The logger will be initialized if it has not been initialized. By default a StreamHandler will be added. If log_file is specified, a FileHandler will also be added. The name of the root logger is the top-level package name, e.g., “mmpose”.

参数
  • log_file (str | None) – The log filename. If specified, a FileHandler will be added to the root logger.

  • log_level (int) – The root logger level. Note that only the process of rank 0 is affected, while other processes will set the level to “Error” and be silent most of the time.

返回

The root logger.

返回类型

logging.Logger

mmocr.utils.is_not_png(img_file)[源代码]

Check img_file is not png image.

参数

img_file (str) – The input image file name

返回

The bool flag indicating whether it is not png

mmocr.utils.is_on_same_line(box_a, box_b, min_y_overlap_ratio=0.8)[源代码]

Check if two boxes are on the same line by their y-axis coordinates.

Two boxes are on the same line if they overlap vertically, and the length of the overlapping line segment is greater than min_y_overlap_ratio * the height of either of the boxes.

参数
  • box_a (list), box_b (list) – Two bounding boxes to be checked

  • min_y_overlap_ratio (float) – The minimum vertical overlapping ratio allowed for boxes in the same line

返回

The bool flag indicating if they are on the same line

mmocr.utils.list_from_file(filename, encoding='utf-8')[源代码]

Load a text file and parse the content as a list of strings. The trailing “r” and “n” of each line will be removed.

注解

This will be replaced by mmcv’s version after it supports encoding.

参数
  • filename (str) – Filename.

  • encoding (str) – Encoding used to open the file. Default utf-8.

返回

A list of strings.

返回类型

list[str]

mmocr.utils.list_to_file(filename, lines)[源代码]

Write a list of strings to a text file.

参数
  • filename (str) – The output filename. It will be created/overwritten.

  • lines (list(str)) – Data to be written.

mmocr.utils.revert_sync_batchnorm(module)[源代码]

Helper function to convert all SyncBatchNorm layers in the model to BatchNormXd layers.

Adapted from @kapily’s work: (https://github.com/pytorch/pytorch/issues/41081#issuecomment-783961547)

参数

module (nn.Module) – The module containing SyncBatchNorm layers.

返回

The converted module with BatchNormXd layers.

返回类型

module_output

mmocr.utils.stitch_boxes_into_lines(boxes, max_x_dist=10, min_y_overlap_ratio=0.8)[源代码]

Stitch fragmented boxes of words into lines.

Note: part of its logic is inspired by @Johndirr (https://github.com/faustomorales/keras-ocr/issues/22)

参数
  • boxes (list) – List of ocr results to be stitched

  • max_x_dist (int) – The maximum horizontal distance between the closest edges of neighboring boxes in the same line

  • min_y_overlap_ratio (float) – The minimum vertical overlapping ratio allowed for any pairs of neighboring boxes in the same line

返回

List of merged boxes and texts

返回类型

merged_boxes(list[dict])

mmocr.models

common_backbones

class mmocr.models.common.backbones.UNet(in_channels=3, base_channels=64, num_stages=5, strides=(1, 1, 1, 1, 1), enc_num_convs=(2, 2, 2, 2, 2), dec_num_convs=(2, 2, 2, 2), downsamples=(True, True, True, True), enc_dilations=(1, 1, 1, 1, 1), dec_dilations=(1, 1, 1, 1), with_cp=False, conv_cfg=None, norm_cfg={'type': 'BN'}, act_cfg={'type': 'ReLU'}, upsample_cfg={'type': 'InterpConv'}, norm_eval=False, dcn=None, plugins=None, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Constant', 'layer': ['_BatchNorm', 'GroupNorm'], 'val': 1}])[源代码]

UNet backbone. U-Net: Convolutional Networks for Biomedical Image Segmentation. https://arxiv.org/pdf/1505.04597.pdf

参数
  • in_channels (int) – Number of input image channels. Default” 3.

  • base_channels (int) – Number of base channels of each stage. The output channels of the first stage. Default: 64.

  • num_stages (int) – Number of stages in encoder, normally 5. Default: 5.

  • strides (Sequence[int 1 | 2]) – Strides of each stage in encoder. len(strides) is equal to num_stages. Normally the stride of the first stage in encoder is 1. If strides[i]=2, it uses stride convolution to downsample in the correspondence encoder stage. Default: (1, 1, 1, 1, 1).

  • enc_num_convs (Sequence[int]) – Number of convolutional layers in the convolution block of the correspondence encoder stage. Default: (2, 2, 2, 2, 2).

  • dec_num_convs (Sequence[int]) – Number of convolutional layers in the convolution block of the correspondence decoder stage. Default: (2, 2, 2, 2).

  • downsamples (Sequence[int]) – Whether use MaxPool to downsample the feature map after the first stage of encoder (stages: [1, num_stages)). If the correspondence encoder stage use stride convolution (strides[i]=2), it will never use MaxPool to downsample, even downsamples[i-1]=True. Default: (True, True, True, True).

  • enc_dilations (Sequence[int]) – Dilation rate of each stage in encoder. Default: (1, 1, 1, 1, 1).

  • dec_dilations (Sequence[int]) – Dilation rate of each stage in decoder. Default: (1, 1, 1, 1).

  • with_cp (bool) – Use checkpoint or not. Using checkpoint will save some memory while slowing down the training speed. Default: False.

  • conv_cfg (dict | None) – Config dict for convolution layer. Default: None.

  • norm_cfg (dict | None) – Config dict for normalization layer. Default: dict(type=’BN’).

  • act_cfg (dict | None) – Config dict for activation layer in ConvModule. Default: dict(type=’ReLU’).

  • upsample_cfg (dict) – The upsample config of the upsample module in decoder. Default: dict(type=’InterpConv’).

  • norm_eval (bool) – Whether to set norm layers to eval mode, namely, freeze running stats (mean and var). Note: Effect on Batch Norm and its variants only. Default: False.

  • dcn (bool) – Use deformable convolution in convolutional layer or not. Default: None.

  • plugins (dict) – plugins for convolutional layers. Default: None.

Notice:

The input image size should be divisible by the whole downsample rate of the encoder. More detail of the whole downsample rate can be found in UNet._check_input_divisible.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

train(mode=True)[源代码]

Convert the model into training mode while keep normalization layer freezed.

class mmocr.models.common.losses.DiceLoss(eps=1e-06)[源代码]
forward(pred, target, mask=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.common.losses.FocalLoss(gamma=2, weight=None, ignore_index=- 100)[源代码]

Multi-class Focal loss implementation.

参数
  • gamma (float) – The larger the gamma, the smaller the loss weight of easier samples.

  • weight (float) – A manual rescaling weight given to each class.

  • ignore_index (int) – Specifies a target value that is ignored and does not contribute to the input gradient.

forward(input, target)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textdet_dense_heads

class mmocr.models.textdet.dense_heads.DBHead(in_channels, with_bias=False, decoding_type='db', text_repr_type='poly', downsample_ratio=1.0, loss={'type': 'DBLoss'}, train_cfg=None, test_cfg=None, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv'}, {'type': 'Constant', 'layer': 'BatchNorm', 'val': 1.0, 'bias': 0.0001}])[源代码]

The class for DBNet head.

This was partially adapted from https://github.com/MhLiao/DB

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.dense_heads.DRRGHead(in_channels, k_at_hops=(8, 4), num_adjacent_linkages=3, node_geo_feat_len=120, pooling_scale=1.0, pooling_output_size=(4, 3), nms_thr=0.3, min_width=8.0, max_width=24.0, comp_shrink_ratio=1.03, comp_ratio=0.4, comp_score_thr=0.3, text_region_thr=0.2, center_region_thr=0.2, center_region_area_thr=50, local_graph_thr=0.7, link_thr=0.85, loss={'type': 'DRRGLoss'}, train_cfg=None, test_cfg=None, init_cfg={'mean': 0, 'override': {'name': 'out_conv'}, 'std': 0.01, 'type': 'Normal'})[源代码]

The class for DRRG head: Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection.

[https://arxiv.org/abs/2003.07493]

参数
  • k_at_hops (tuple(int)) – The number of i-hop neighbors, i = 1, 2.

  • num_adjacent_linkages (int) – The number of linkages when constructing adjacent matrix.

  • node_geo_feat_len (int) – The length of embedded geometric feature vector of a component.

  • pooling_scale (float) – The spatial scale of rotated RoI-Align.

  • pooling_output_size (tuple(int)) – The output size of RRoI-Aligning.

  • nms_thr (float) – The locality-aware NMS threshold of text components.

  • min_width (float) – The minimum width of text components.

  • max_width (float) – The maximum width of text components.

  • comp_shrink_ratio (float) – The shrink ratio of text components.

  • comp_ratio (float) – The reciprocal of aspect ratio of text components.

  • comp_score_thr (float) – The score threshold of text components.

  • text_region_thr (float) – The threshold for text region probability map.

  • center_region_thr (float) – The threshold for text center region probability map.

  • center_region_area_thr (int) – The threshold for filtering small-sized text center region.

  • local_graph_thr (float) – The threshold to filter identical local graphs.

  • link_thr (float) – The threshold for connected components search.

forward(inputs, gt_comp_attribs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_boundary(edges, scores, text_comps, img_metas, rescale)[源代码]

Compute text boundaries via post processing.

参数
  • edges (ndarray) – The edge array of shape N * 2, each row is a pair of text component indices that makes up an edge in graph.

  • scores (ndarray) – The edge score array.

  • text_comps (ndarray) – The text components.

  • img_metas (list[dict]) – The image meta infos.

  • rescale (bool) – Rescale boundaries to the original image resolution.

返回

The result dict.

返回类型

results (dict)

class mmocr.models.textdet.dense_heads.FCEHead(in_channels, scales, fourier_degree=5, num_sample=50, num_reconstr_points=50, decoding_type='fcenet', loss={'type': 'FCELoss'}, score_thr=0.3, nms_thr=0.1, alpha=1.0, beta=1.0, text_repr_type='poly', train_cfg=None, test_cfg=None, init_cfg={'mean': 0, 'override': [{'name': 'out_conv_cls'}, {'name': 'out_conv_reg'}], 'std': 0.01, 'type': 'Normal'})[源代码]

The class for implementing FCENet head. FCENet(CVPR2021): Fourier Contour Embedding for Arbitrary-shaped Text Detection.

[https://arxiv.org/abs/2104.10442]

参数
  • in_channels (int) – The number of input channels.

  • scales (list[int]) – The scale of each layer.

  • fourier_degree (int) – The maximum Fourier transform degree k.

  • num_sample (int) – The sampling points number of regression loss. If it is too small, FCEnet tends to be overfitting.

  • score_thr (float) – The threshold to filter out the final candidates.

  • nms_thr (float) – The threshold of nms.

  • alpha (float) – The parameter to calculate final scores. Score_{final} = (Score_{text region} ^ alpha) * (Score{text center region} ^ beta)

  • beta (float) – The parameter to calculate final scores.

forward(feats)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

get_boundary(score_maps, img_metas, rescale)[源代码]

Compute text boundaries via post processing.

参数
  • score_maps (Tensor) – The text score map.

  • img_metas (dict) – The image meta info.

  • rescale (bool) – Rescale boundaries to the original image resolution if true, and keep the score_maps resolution if false.

返回

The result dict.

返回类型

results (dict)

class mmocr.models.textdet.dense_heads.HeadMixin[源代码]

The head minxin for dbnet and pannet heads.

get_boundary(score_maps, img_metas, rescale)[源代码]

Compute text boundaries via post processing.

参数
  • score_maps (Tensor) – The text score map.

  • img_metas (dict) – The image meta info.

  • rescale (bool) – Rescale boundaries to the original image resolution if true, and keep the score_maps resolution if false.

返回

The result dict.

返回类型

results (dict)

loss(pred_maps, **kwargs)[源代码]

Compute the loss for text detection.

参数

pred_maps (tensor) – The input score maps of NxCxHxW.

返回

The dict for losses.

返回类型

losses (dict)

resize_boundary(boundaries, scale_factor)[源代码]

Rescale boundaries via scale_factor.

参数
  • boundaries (list[list[float]]) – The boundary list. Each boundary

  • size 2k+1 with k>=4. (with) –

  • scale_factor (ndarray) – The scale factor of size (4,).

返回

The scaled boundaries.

返回类型

boundaries (list[list[float]])

class mmocr.models.textdet.dense_heads.PANHead(in_channels, out_channels, text_repr_type='poly', downsample_ratio=0.25, loss={'type': 'PANLoss'}, train_cfg=None, test_cfg=None, init_cfg={'mean': 0, 'override': {'name': 'out_conv'}, 'std': 0.01, 'type': 'Normal'})[源代码]

The class for PANet head.

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.dense_heads.PSEHead(in_channels, out_channels, text_repr_type='poly', downsample_ratio=0.25, loss={'type': 'PSELoss'}, train_cfg=None, test_cfg=None, init_cfg=None)[源代码]

The class for PANet head.

class mmocr.models.textdet.dense_heads.TextSnakeHead(in_channels, decoding_type='textsnake', text_repr_type='poly', loss={'type': 'TextSnakeLoss'}, train_cfg=None, test_cfg=None, init_cfg={'mean': 0, 'override': {'name': 'out_conv'}, 'std': 0.01, 'type': 'Normal'})[源代码]

The class for TextSnake head: TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes.

[https://arxiv.org/abs/1807.01544]

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textdet_necks

class mmocr.models.textdet.necks.FPEM_FFM(in_channels, conv_out=128, fpem_repeat=2, align_corners=False, init_cfg={'distribution': 'uniform', 'layer': 'Conv2d', 'type': 'Xavier'})[源代码]

This code is from https://github.com/WenmuZhou/PAN.pytorch.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.necks.FPNC(in_channels, lateral_channels=256, out_channels=64, bias_on_lateral=False, bn_re_on_lateral=False, bias_on_smooth=False, bn_re_on_smooth=False, conv_after_concat=False, init_cfg=None)[源代码]

FPN-like fusion module in Real-time Scene Text Detection with Differentiable Binarization.

This was partially adapted from https://github.com/MhLiao/DB and https://github.com/WenmuZhou/DBNet.pytorch

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.necks.FPNF(in_channels=[256, 512, 1024, 2048], out_channels=256, fusion_type='concat', upsample_ratio=1, init_cfg={'distribution': 'uniform', 'layer': 'Conv2d', 'type': 'Xavier'})[源代码]

FPN-like fusion module in Shape Robust Text Detection with Progressive Scale Expansion Network.

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.necks.FPN_UNet(in_channels, out_channels, init_cfg={'distribution': 'uniform', 'layer': ['Conv2d', 'ConvTranspose2d'], 'type': 'Xavier'})[源代码]

The class for implementing DRRG and TextSnake U-Net-like FPN.

DRRG: Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection [https://arxiv.org/abs/2003.07493]. TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes [https://arxiv.org/abs/1807.01544].

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textdet_detectors

class mmocr.models.textdet.detectors.DBNet(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing DBNet text detector: Real-time Scene Text Detection with Differentiable Binarization.

[https://arxiv.org/abs/1911.08947].

class mmocr.models.textdet.detectors.DRRG(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing DRRG text detector. Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection.

[https://arxiv.org/abs/2003.07493]

forward_train(img, img_metas, **kwargs)[源代码]
参数
  • img (Tensor) – Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – A List of image info dict where each dict has: ‘img_shape’, ‘scale_factor’, ‘flip’, and may also contain ‘filename’, ‘ori_shape’, ‘pad_shape’, and ‘img_norm_cfg’. For details of the values of these keys see mmdet.datasets.pipelines.Collect.

返回

A dictionary of loss components.

返回类型

dict[str, Tensor]

simple_test(img, img_metas, rescale=False)[源代码]

Test function without test-time augmentation.

参数
  • img (torch.Tensor) – Images with shape (N, C, H, W).

  • img_metas (list[dict]) – List of image information.

  • rescale (bool, optional) – Whether to rescale the results. Defaults to False.

返回

BBox results of each image and classes.

The outer list corresponds to each image. The inner list corresponds to each class.

返回类型

list[list[np.ndarray]]

class mmocr.models.textdet.detectors.FCENet(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing FCENet text detector FCENet(CVPR2021): Fourier Contour Embedding for Arbitrary-shaped Text

Detection

[https://arxiv.org/abs/2104.10442]

simple_test(img, img_metas, rescale=False)[源代码]

Test function without test-time augmentation.

参数
  • img (torch.Tensor) – Images with shape (N, C, H, W).

  • img_metas (list[dict]) – List of image information.

  • rescale (bool, optional) – Whether to rescale the results. Defaults to False.

返回

BBox results of each image and classes.

The outer list corresponds to each image. The inner list corresponds to each class.

返回类型

list[list[np.ndarray]]

class mmocr.models.textdet.detectors.OCRMaskRCNN(backbone, rpn_head, roi_head, train_cfg, test_cfg, neck=None, pretrained=None, text_repr_type='quad', show_score=False, init_cfg=None)[源代码]

Mask RCNN tailored for OCR.

simple_test(img, img_metas, proposals=None, rescale=False)[源代码]

Test without augmentation.

class mmocr.models.textdet.detectors.PANet(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing PANet text detector:

Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network [https://arxiv.org/abs/1908.05900].

class mmocr.models.textdet.detectors.PSENet(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing PSENet text detector: Shape Robust Text Detection with Progressive Scale Expansion Network.

[https://arxiv.org/abs/1806.02559].

class mmocr.models.textdet.detectors.SingleStageTextDetector(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, init_cfg=None)[源代码]

The class for implementing single stage text detector.

It is the parent class of PANet, PSENet, and DBNet.

forward_train(img, img_metas, **kwargs)[源代码]
参数
  • img (Tensor) – Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – A list of image info dict where each dict has: ‘img_shape’, ‘scale_factor’, ‘flip’, and may also contain ‘filename’, ‘ori_shape’, ‘pad_shape’, and ‘img_norm_cfg’. For details on the values of these keys, see mmdet.datasets.pipelines.Collect.

返回

A dictionary of loss components.

返回类型

dict[str, Tensor]

simple_test(img, img_metas, rescale=False)[源代码]

Test function without test-time augmentation.

参数
  • img (torch.Tensor) – Images with shape (N, C, H, W).

  • img_metas (list[dict]) – List of image information.

  • rescale (bool, optional) – Whether to rescale the results. Defaults to False.

返回

BBox results of each image and classes.

The outer list corresponds to each image. The inner list corresponds to each class.

返回类型

list[list[np.ndarray]]

class mmocr.models.textdet.detectors.TextDetectorMixin(show_score)[源代码]

The class for implementing text detector auxiliary methods.

get_boundary(results)[源代码]

Convert segmentation into text boundaries.

参数

results (tuple) – The result tuple. The first element is segmentation while the second is its scores.

返回

A result dict containing ‘boundary_result’.

返回类型

results (dict)

show_result(img, result, score_thr=0.5, bbox_color='green', text_color='green', thickness=1, font_scale=0.5, win_name='', show=False, wait_time=0, out_file=None)[源代码]

Draw result over img.

参数
  • img (str or Tensor) – The image to be displayed.

  • result (dict) – The results to draw over img.

  • score_thr (float, optional) – Minimum score of bboxes to be shown. Default: 0.3.

  • bbox_color (str or tuple or Color) – Color of bbox lines.

  • text_color (str or tuple or Color) – Color of texts.

  • thickness (int) – Thickness of lines.

  • font_scale (float) – Font scales of texts.

  • win_name (str) – The window name.

  • wait_time (int) – Value of waitKey param. Default: 0.

  • show (bool) – Whether to show the image. Default: False.

  • out_file (str or None) – The filename to write the image. Default: None.imshow_pred_boundary`

class mmocr.models.textdet.detectors.TextSnake(backbone, neck, bbox_head, train_cfg=None, test_cfg=None, pretrained=None, show_score=False, init_cfg=None)[源代码]

The class for implementing TextSnake text detector: TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes.

[https://arxiv.org/abs/1807.01544]

textdet_losses

class mmocr.models.textdet.losses.DBLoss(alpha=1, beta=1, reduction='mean', negative_ratio=3.0, eps=1e-06, bbce_loss=False)[源代码]

The class for implementing DBNet loss.

This is partially adapted from https://github.com/MhLiao/DB.

bitmasks2tensor(bitmasks, target_sz)[源代码]

Convert Bitmasks to tensor.

参数
  • bitmasks (list[BitMasks]) – The BitMasks list. Each item is for one img.

  • target_sz (tuple(int, int)) – The target tensor size of KxHxW with K being the number of kernels.

Returns
result_tensors (list[tensor]): The list of kernel tensors. Each

element is for one kernel level.

forward(preds, downsample_ratio, gt_shrink, gt_shrink_mask, gt_thr, gt_thr_mask)[源代码]

Compute DBNet loss.

参数
  • preds (tensor) – The output tensor with size of Nx3xHxW.

  • downsample_ratio (float) – The downsample ratio for the ground truths.

  • gt_shrink (list[BitmapMasks]) – The mask list with each element being the shrunk text mask for one img.

  • gt_shrink_mask (list[BitmapMasks]) – The effective mask list with each element being the shrunk effective mask for one img.

  • gt_thr (list[BitmapMasks]) – The mask list with each element being the threshold text mask for one img.

  • gt_thr_mask (list[BitmapMasks]) – The effective mask list with each element being the threshold effective mask for one img.

返回

The dict for dbnet losses with loss_prob,

loss_db and loss_thresh.

返回类型

results(dict)

class mmocr.models.textdet.losses.DRRGLoss(ohem_ratio=3.0)[源代码]

The class for implementing DRRG loss: Deep Relational Reasoning Graph Network for Arbitrary Shape Text Detection.

[https://arxiv.org/abs/1908.05900] This is partially adapted from https://github.com/GXYM/DRRG licensed under the MIT license.

bitmasks2tensor(bitmasks, target_sz)[源代码]

Convert Bitmasks to tensor.

参数
  • bitmasks (list[BitmapMasks]) – The BitmapMasks list. Each item is for one img.

  • target_sz (tuple(int, int)) – The target tensor size HxW.

Returns
results (list[tensor]): The list of kernel tensors. Each

element is for one kernel level.

forward(preds, downsample_ratio, gt_text_mask, gt_center_region_mask, gt_mask, gt_top_height_map, gt_bot_height_map, gt_sin_map, gt_cos_map)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textdet.losses.FCELoss(fourier_degree, num_sample, ohem_ratio=3.0)[源代码]

The class for implementing FCENet loss FCENet(CVPR2021): Fourier Contour Embedding for Arbitrary-shaped

Text Detection

[https://arxiv.org/abs/2104.10442]

参数
  • fourier_degree (int) – The maximum Fourier transform degree k.

  • num_sample (int) – The sampling points number of regression loss. If it is too small, fcenet tends to be overfitting.

  • ohem_ratio (float) – the negative/positive ratio in OHEM.

forward(preds, _, p3_maps, p4_maps, p5_maps)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

fourier2poly(real_maps, imag_maps)[源代码]

Transform Fourier coefficient maps to polygon maps.

参数
  • real_maps (tensor) – A map composed of the real parts of the Fourier coefficients, whose shape is (-1, 2k+1)

  • imag_maps (tensor) – A map composed of the imag parts of the Fourier coefficients, whose shape is (-1, 2k+1)

Returns
x_maps (tensor): A map composed of the x value of the polygon

represented by n sample points (xn, yn), whose shape is (-1, n)

y_maps (tensor): A map composed of the y value of the polygon

represented by n sample points (xn, yn), whose shape is (-1, n)

class mmocr.models.textdet.losses.PANLoss(alpha=0.5, beta=0.25, delta_aggregation=0.5, delta_discrimination=3, ohem_ratio=3, reduction='mean', speedup_bbox_thr=- 1)[源代码]

The class for implementing PANet loss: Efficient and Accurate Arbitrary- Shaped Text Detection with Pixel Aggregation Network.

[https://arxiv.org/abs/1908.05900]. This was partially adapted from https://github.com/WenmuZhou/PAN.pytorch

aggregation_discrimination_loss(gt_texts, gt_kernels, inst_embeds)[源代码]

Compute the aggregation and discrimnative losses.

参数
  • gt_texts (tensor) – The ground truth text mask of size Nx1xHxW.

  • gt_kernels (tensor) – The ground truth text kernel mask of size Nx1xHxW.

  • inst_embeds (tensor) – The text instance embedding tensor of size Nx4xHxW.

返回

The aggregation loss before reduction. loss_discrs (tensor): The discriminative loss before reduction.

返回类型

loss_aggrs (tensor)

bitmasks2tensor(bitmasks, target_sz)[源代码]

Convert Bitmasks to tensor.

参数
  • bitmasks (list[BitmapMasks]) – The BitmapMasks list. Each item is for one img.

  • target_sz (tuple(int, int)) – The target tensor size HxW.

Returns
results (list[tensor]): The list of kernel tensors. Each

element is for one kernel level.

forward(preds, downsample_ratio, gt_kernels, gt_mask)[源代码]

Compute PANet loss.

参数
  • preds (tensor) – The output tensor with size of Nx6xHxW.

  • gt_kernels (list[BitmapMasks]) – The kernel list with each element being the text kernel mask for one img.

  • gt_mask (list[BitmapMasks]) – The effective mask list with each element being the effective mask for one img.

  • downsample_ratio (float) – The downsample ratio between preds and the input img.

返回

The loss dictionary.

返回类型

results (dict)

ohem_batch(text_scores, gt_texts, gt_mask)[源代码]

OHEM sampling for a batch of imgs.

参数
  • text_scores (Tensor) – The text scores of size NxHxW.

  • gt_texts (Tensor) – The gt text masks of size NxHxW.

  • gt_mask (Tensor) – The gt effective mask of size NxHxW.

返回

The sampled mask of size NxHxW.

返回类型

sampled_masks (Tensor)

ohem_img(text_score, gt_text, gt_mask)[源代码]

Sample the top-k maximal negative samples and all positive samples.

参数
  • text_score (Tensor) – The text score with size of HxW.

  • gt_text (Tensor) – The ground truth text mask of HxW.

  • gt_mask (Tensor) – The effective region mask of HxW.

返回

The sampled pixel mask of size HxW.

返回类型

sampled_mask (Tensor)

class mmocr.models.textdet.losses.PSELoss(alpha=0.7, ohem_ratio=3, reduction='mean', kernel_sample_type='adaptive')[源代码]

The class for implementing PSENet loss: Shape Robust Text Detection with Progressive Scale Expansion Network [https://arxiv.org/abs/1806.02559].

This is partially adapted from https://github.com/whai362/PSENet.

forward(score_maps, downsample_ratio, gt_kernels, gt_mask)[源代码]

Compute PSENet loss.

参数
  • score_maps (tensor) – The output tensor with size of Nx6xHxW.

  • gt_kernels (list[BitmapMasks]) – The kernel list with each element being the text kernel mask for one img.

  • gt_mask (list[BitmapMasks]) – The effective mask list with each element being the effective mask for one img.

  • downsample_ratio (float) – The downsample ratio between score_maps and the input img.

返回

The loss.

返回类型

results (dict)

class mmocr.models.textdet.losses.TextSnakeLoss(ohem_ratio=3.0)[源代码]

The class for implementing TextSnake loss: TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes [https://arxiv.org/abs/1807.01544]. This is partially adapted from https://github.com/princewang1994/TextSnake.pytorch.

bitmasks2tensor(bitmasks, target_sz)[源代码]

Convert Bitmasks to tensor.

参数
  • bitmasks (list[BitmapMasks]) – The BitmapMasks list. Each item is for one img.

  • target_sz (tuple(int, int)) – The target tensor size HxW.

Returns
results (list[tensor]): The list of kernel tensors. Each

element is for one kernel level.

forward(pred_maps, downsample_ratio, gt_text_mask, gt_center_region_mask, gt_mask, gt_radius_map, gt_sin_map, gt_cos_map)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textdet_postprocess

textrecog_recognizer

class mmocr.models.textrecog.recognizer.BaseRecognizer(init_cfg=None)[源代码]

Base class for text recognition.

abstract aug_test(imgs, img_metas, **kwargs)[源代码]

Test function with test time augmentation.

参数
  • imgs (list[tensor]) – Tensor should have shape NxCxHxW, which contains all images in the batch.

  • img_metas (list[list[dict]]) – The metadata of images.

abstract extract_feat(imgs)[源代码]

Extract features from images.

forward(img, img_metas, return_loss=True, **kwargs)[源代码]

Calls either forward_train() or forward_test() depending on whether return_loss is True.

Note that img and img_meta are single-nested (i.e. tensor and list[dict]).

forward_test(imgs, img_metas, **kwargs)[源代码]
参数
  • imgs (tensor | list[tensor]) – Tensor should have shape NxCxHxW, which contains all images in the batch.

  • img_metas (list[dict] | list[list[dict]]) – The outer list indicates images in a batch.

abstract forward_train(imgs, img_metas, **kwargs)[源代码]
参数
  • img (tensor) – tensors with shape (N, C, H, W). Typically should be mean centered and std scaled.

  • img_metas (list[dict]) – List of image info dict where each dict has: ‘img_shape’, ‘scale_factor’, ‘flip’, and may also contain ‘filename’, ‘ori_shape’, ‘pad_shape’, and ‘img_norm_cfg’. For details of the values of these keys, see mmdet.datasets.pipelines.Collect.

  • kwargs (keyword arguments) – Specific to concrete implementation.

show_result(img, result, gt_label='', win_name='', show=False, wait_time=0, out_file=None, **kwargs)[源代码]

Draw result on img.

参数
  • img (str or tensor) – The image to be displayed.

  • result (dict) – The results to draw on img.

  • gt_label (str) – Ground truth label of img.

  • win_name (str) – The window name.

  • wait_time (int) – Value of waitKey param. Default: 0.

  • show (bool) – Whether to show the image. Default: False.

  • out_file (str or None) – The output filename. Default: None.

返回

Only if not show or out_file.

返回类型

img (tensor)

train_step(data, optimizer)[源代码]

The iteration step during training.

This method defines an iteration step during training, except for the back propagation and optimizer update, which are done by an optimizer hook. Note that in some complicated cases or models (e.g. GAN), the whole process (including the back propagation and optimizer update) is also defined by this method.

参数
  • data (dict) – The outputs of dataloader.

  • optimizer (torch.optim.Optimizer | dict) – The optimizer of runner is passed to train_step(). This argument is unused and reserved.

返回

It should contain at least 3 keys: loss, log_vars,

num_samples.

  • loss is a tensor for back propagation, which is a

weighted sum of multiple losses. - log_vars contains all the variables to be sent to the logger. - num_samples indicates the batch size used for averaging the logs (Note: for the DDP model, num_samples refers to the batch size for each GPU).

返回类型

dict

val_step(data, optimizer)[源代码]

The iteration step during validation.

This method shares the same signature as train_step(), but is used during val epochs. Note that the evaluation after training epochs is not implemented by this method, but by an evaluation hook.

class mmocr.models.textrecog.recognizer.CRNNNet(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

CTC-loss based recognizer.

class mmocr.models.textrecog.recognizer.EncodeDecodeRecognizer(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

Base class for encode-decode recognizer.

aug_test(imgs, img_metas, **kwargs)[源代码]

Test function as well as time augmentation.

参数
  • imgs (list[tensor]) – Tensor should have shape NxCxHxW, which contains all images in the batch.

  • img_metas (list[list[dict]]) – The metadata of images.

extract_feat(img)[源代码]

Directly extract features from the backbone.

forward_train(img, img_metas)[源代码]
参数
  • img (tensor) – Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – A list of image info dict where each dict contains: ‘img_shape’, ‘filename’, and may also contain ‘ori_shape’, and ‘img_norm_cfg’. For details on the values of these keys see mmdet.datasets.pipelines.Collect.

返回

A dictionary of loss components.

返回类型

dict[str, tensor]

simple_test(img, img_metas, **kwargs)[源代码]

Test function with test time augmentation.

参数
  • imgs (torch.Tensor) – Image input tensor.

  • img_metas (list[dict]) – List of image information.

返回

Text label result of each image.

返回类型

list[str]

class mmocr.models.textrecog.recognizer.NRTR(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

Implementation of NRTR

class mmocr.models.textrecog.recognizer.RobustScanner(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

Implementation of `RobustScanner.

<https://arxiv.org/pdf/2007.07542.pdf>

class mmocr.models.textrecog.recognizer.SARNet(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

Implementation of SAR

class mmocr.models.textrecog.recognizer.SATRN(preprocessor=None, backbone=None, encoder=None, decoder=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, max_seq_len=40, pretrained=None, init_cfg=None)[源代码]

Implementation of SATRN

class mmocr.models.textrecog.recognizer.SegRecognizer(preprocessor=None, backbone=None, neck=None, head=None, loss=None, label_convertor=None, train_cfg=None, test_cfg=None, pretrained=None, init_cfg=None)[源代码]

Base class for segmentation based recognizer.

aug_test(imgs, img_metas, **kwargs)[源代码]

Test function with test time augmentation.

参数
  • imgs (list[tensor]) – Tensor should have shape NxCxHxW, which contains all images in the batch.

  • img_metas (list[list[dict]]) – The metadata of images.

extract_feat(img)[源代码]

Directly extract features from the backbone.

forward_train(img, img_metas, gt_kernels=None)[源代码]
参数
  • img (tensor) – Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – A list of image info dict where each dict contains: ‘img_shape’, ‘filename’, and may also contain ‘ori_shape’, and ‘img_norm_cfg’. For details on the values of these keys see mmdet.datasets.pipelines.Collect.

返回

A dictionary of loss components.

返回类型

dict[str, tensor]

simple_test(img, img_metas, **kwargs)[源代码]

Test function without test time augmentation.

参数
  • imgs (torch.Tensor) – Image input tensor.

  • img_metas (list[dict]) – List of image information.

返回

Text label result of each image.

返回类型

list[str]

textrecog_backbones

class mmocr.models.textrecog.backbones.NRTRModalityTransform(input_channels=3, input_height=32, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.ResNet31OCR(base_channels=3, layers=[1, 2, 5, 3], channels=[64, 128, 256, 256, 512, 512, 512], out_indices=None, stage4_pool_cfg={'kernel_size': (2, 1), 'stride': (2, 1)}, last_stage_pool=False, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement ResNet backbone for text recognition, modified from

ResNet

参数
  • base_channels (int) – Number of channels of input image tensor.

  • layers (list[int]) – List of BasicBlock number for each stage.

  • channels (list[int]) – List of out_channels of Conv2d layer.

  • out_indices (None | Sequence[int]) – Indices of output stages.

  • stage4_pool_cfg (dict) – Dictionary to construct and configure pooling layer in stage 4.

  • last_stage_pool (bool) – If True, add MaxPool2d layer to last stage.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.ShallowCNN(input_channels=1, hidden_dim=512, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement Shallow CNN block for SATRN, see

SATRN

参数
  • base_channels (int) – Number of channels of input image tensor.

  • hidden_dim (int) – Size of hidden layers of the model.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.VeryDeepVgg(leaky_relu=True, input_channels=3, init_cfg=[{'type': 'Xavier', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement VGG-VeryDeep backbone for text recognition, modified from

VGG-VeryDeep

参数
  • leaky_relu (bool) – Use leakyRelu or not.

  • input_channels (int) – Number of channels of input image tensor.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textrecog_necks

class mmocr.models.textrecog.necks.FPNOCR(in_channels, out_channels, last_stage_only=True, init_cfg=None)[源代码]

FPN-like Network for segmentation based text recognition.

参数
  • in_channels (list[int]) – Number of input channels for each scale.

  • out_channels (int) – Number of output channels for each scale.

  • last_stage_only (bool) – If True, output last stage only.

forward(inputs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textrecog_heads

class mmocr.models.textrecog.heads.SegHead(in_channels=128, num_classes=37, upsample_param=None, init_cfg=None)[源代码]

Head for segmentation based text recognition.

参数
  • in_channels (int) – Number of input channels.

  • num_classes (int) – Number of output classes.

  • upsample_param (dict | None) – Config dict for interpolation layer. Default: dict(scale_factor=1.0, mode=’nearest’)

forward(out_neck)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textrecog_convertors

class mmocr.models.textrecog.convertors.AttnConvertor(dict_type='DICT90', dict_file=None, dict_list=None, with_unknown=True, max_seq_len=40, lower=False, start_end_same=True, **kwargs)[源代码]

Convert between text, index and tensor for encoder-decoder based pipeline.

参数
  • dict_type (str) – Type of dict, should be one of {‘DICT36’, ‘DICT90’}.

  • dict_file (None|str) – Character dict file path. If not none, higher priority than dict_type.

  • dict_list (None|list[str]) – Character list. If not none, higher priority than dict_type, but lower than dict_file.

  • with_unknown (bool) – If True, add UKN token to class.

  • max_seq_len (int) – Maximum sequence length of label.

  • lower (bool) – If True, convert original string to lower case.

  • start_end_same (bool) – Whether use the same index for start and end token or not. Default: True.

str2tensor(strings)[源代码]

Convert text-string into tensor. :param strings: [‘hello’, ‘world’] :type strings: list[str]

返回

Tensor | list[tensor]):
tensors (list[Tensor]): [torch.Tensor([1,2,3,3,4]),

torch.Tensor([5,4,6,3,7])]

padded_targets (Tensor(bsz * max_seq_len))

返回类型

dict (str

tensor2idx(outputs, img_metas=None)[源代码]

Convert output tensor to text-index :param outputs: model outputs with size: N * T * C :type outputs: tensor :param img_metas: Each dict contains one image info. :type img_metas: list[dict]

返回

[[1,2,3,3,4], [5,4,6,3,7]] scores (list[list[float]]): [[0.9,0.8,0.95,0.97,0.94],

[0.9,0.9,0.98,0.97,0.96]]

返回类型

indexes (list[list[int]])

class mmocr.models.textrecog.convertors.BaseConvertor(dict_type='DICT90', dict_file=None, dict_list=None)[源代码]

Convert between text, index and tensor for text recognize pipeline.

参数
  • dict_type (str) – Type of dict, should be either ‘DICT36’ or ‘DICT90’.

  • dict_file (None|str) – Character dict file path. If not none, the dict_file is of higher priority than dict_type.

  • dict_list (None|list[str]) – Character list. If not none, the list is of higher priority than dict_type, but lower than dict_file.

idx2str(indexes)[源代码]

Convert indexes to text strings.

参数

indexes (list[list[int]]) – [[1,2,3,3,4], [5,4,6,3,7]].

返回

[‘hello’, ‘world’].

返回类型

strings (list[str])

num_classes()[源代码]

Number of output classes.

str2idx(strings)[源代码]

Convert strings to indexes.

参数

strings (list[str]) – [‘hello’, ‘world’].

返回

[[1,2,3,3,4], [5,4,6,3,7]].

返回类型

indexes (list[list[int]])

str2tensor(strings)[源代码]

Convert text-string to input tensor.

参数

strings (list[str]) – [‘hello’, ‘world’].

返回

[torch.Tensor([1,2,3,3,4]),

torch.Tensor([5,4,6,3,7])].

返回类型

tensors (list[torch.Tensor])

tensor2idx(output)[源代码]

Convert model output tensor to character indexes and scores. :param output: The model outputs with size: N * T * C :type output: tensor

返回

[[1,2,3,3,4], [5,4,6,3,7]]. scores (list[list[float]]): [[0.9,0.8,0.95,0.97,0.94],

[0.9,0.9,0.98,0.97,0.96]].

返回类型

indexes (list[list[int]])

class mmocr.models.textrecog.convertors.CTCConvertor(dict_type='DICT90', dict_file=None, dict_list=None, with_unknown=True, lower=False, **kwargs)[源代码]

Convert between text, index and tensor for CTC loss-based pipeline.

参数
  • dict_type (str) – Type of dict, should be either ‘DICT36’ or ‘DICT90’.

  • dict_file (None|str) – Character dict file path. If not none, the file is of higher priority than dict_type.

  • dict_list (None|list[str]) – Character list. If not none, the list is of higher priority than dict_type, but lower than dict_file.

  • with_unknown (bool) – If True, add UKN token to class.

  • lower (bool) – If True, convert original string to lower case.

str2tensor(strings)[源代码]

Convert text-string to ctc-loss input tensor.

参数

strings (list[str]) – [‘hello’, ‘world’].

返回

tensor | list[tensor]):
tensors (list[tensor]): [torch.Tensor([1,2,3,3,4]),

torch.Tensor([5,4,6,3,7])].

flatten_targets (tensor): torch.Tensor([1,2,3,3,4,5,4,6,3,7]). target_lengths (tensor): torch.IntTensot([5,5]).

返回类型

dict (str

tensor2idx(output, img_metas, topk=1, return_topk=False)[源代码]

Convert model output tensor to index-list. :param output: The model outputs with size: N * T * C. :type output: tensor :param img_metas: Each dict contains one image info. :type img_metas: list[dict] :param topk: The highest k classes to be returned. :type topk: int :param return_topk: Whether to return topk or just top1. :type return_topk: bool

返回

[[1,2,3,3,4], [5,4,6,3,7]]. scores (list[list[float]]): [[0.9,0.8,0.95,0.97,0.94],

[0.9,0.9,0.98,0.97,0.96]] (

indexes_topk (list[list[list[int]->len=topk]]): scores_topk (list[list[list[float]->len=topk]])

).

返回类型

indexes (list[list[int]])

class mmocr.models.textrecog.convertors.SegConvertor(dict_type='DICT36', dict_file=None, dict_list=None, with_unknown=True, lower=False, **kwargs)[源代码]

Convert between text, index and tensor for segmentation based pipeline.

参数
  • dict_type (str) – Type of dict, should be either ‘DICT36’ or ‘DICT90’.

  • dict_file (None|str) – Character dict file path. If not none, the file is of higher priority than dict_type.

  • dict_list (None|list[str]) – Character list. If not none, the list

  • of higher priority than dict_type (is) –

  • lower than dict_file. (but) –

  • with_unknown (bool) – If True, add UKN token to class.

  • lower (bool) – If True, convert original string to lower case.

tensor2str(output, img_metas=None)[源代码]

Convert model output tensor to string labels. :param output: Model outputs with size: N * C * H * W :type output: tensor :param img_metas: Each dict contains one image info. :type img_metas: list[dict]

返回

Decoded text labels. scores (list[list[float]]): Decoded chars scores.

返回类型

texts (list[str])

textrecog_encoders

class mmocr.models.textrecog.encoders.BaseEncoder(init_cfg=None)[源代码]

Base Encoder class for text recognition.

forward(feat, **kwargs)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.encoders.ChannelReductionEncoder(in_channels, out_channels, init_cfg={'layer': 'Conv2d', 'type': 'Xavier'})[源代码]
forward(feat, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.encoders.SAREncoder(enc_bi_rnn=False, enc_do_rnn=0.0, enc_gru=False, d_model=512, d_enc=512, mask=True, init_cfg=[{'type': 'Xavier', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}], **kwargs)[源代码]

Implementation of encoder module in `SAR.

<https://arxiv.org/abs/1811.00751>`_

参数
  • enc_bi_rnn (bool) – If True, use bidirectional RNN in encoder.

  • enc_do_rnn (float) – Dropout probability of RNN layer in encoder.

  • enc_gru (bool) – If True, use GRU, else LSTM in encoder.

  • d_model (int) – Dim of channels from backbone.

  • d_enc (int) – Dim of encoder RNN layer.

  • mask (bool) – If True, mask padding in RNN sequence.

forward(feat, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.encoders.SatrnEncoder(n_layers=12, n_head=8, d_k=64, d_v=64, d_model=512, n_position=100, d_inner=256, dropout=0.1, init_cfg=None, **kwargs)[源代码]

Implement encoder for SATRN, see `SATRN.

<https://arxiv.org/abs/1910.04396>`_.

forward(feat, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.encoders.TFEncoder(n_layers=6, n_head=8, d_k=64, d_v=64, d_model=512, d_inner=256, dropout=0.1, init_cfg=None, **kwargs)[源代码]

Encode 2d feature map to 1d sequence.

forward(feat, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textrecog_decoders

class mmocr.models.textrecog.decoders.BaseDecoder(init_cfg=None, **kwargs)[源代码]

Base decoder class for text recognition.

forward(feat, out_enc, targets_dict=None, img_metas=None, train_mode=True)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.decoders.CRNNDecoder(in_channels=None, num_classes=None, rnn_flag=False, init_cfg={'layer': 'Conv2d', 'type': 'Xavier'}, **kwargs)[源代码]
class mmocr.models.textrecog.decoders.ParallelSARDecoder(num_classes=37, enc_bi_rnn=False, dec_bi_rnn=False, dec_do_rnn=0.0, dec_gru=False, d_model=512, d_enc=512, d_k=64, pred_dropout=0.0, max_seq_len=40, mask=True, start_idx=0, padding_idx=92, pred_concat=False, init_cfg=None, **kwargs)[源代码]

Implementation Parallel Decoder module in `SAR.

<https://arxiv.org/abs/1811.00751>`_

参数
  • number_classes (int) – Output class number.

  • channels (list[int]) – Network layer channels.

  • enc_bi_rnn (bool) – If True, use bidirectional RNN in encoder.

  • dec_bi_rnn (bool) – If True, use bidirectional RNN in decoder.

  • dec_do_rnn (float) – Dropout of RNN layer in decoder.

  • dec_gru (bool) – If True, use GRU, else LSTM in decoder.

  • d_model (int) – Dim of channels from backbone.

  • d_enc (int) – Dim of encoder RNN layer.

  • d_k (int) – Dim of channels of attention module.

  • pred_dropout (float) – Dropout probability of prediction layer.

  • max_seq_len (int) – Maximum sequence length for decoding.

  • mask (bool) – If True, mask padding in feature map.

  • start_idx (int) – Index of start token.

  • padding_idx (int) – Index of padding token.

  • pred_concat (bool) – If True, concat glimpse feature from attention with holistic feature and hidden state.

class mmocr.models.textrecog.decoders.ParallelSARDecoderWithBS(beam_width=5, num_classes=37, enc_bi_rnn=False, dec_bi_rnn=False, dec_do_rnn=0, dec_gru=False, d_model=512, d_enc=512, d_k=64, pred_dropout=0.0, max_seq_len=40, mask=True, start_idx=0, padding_idx=0, pred_concat=False, init_cfg=None, **kwargs)[源代码]

Parallel Decoder module with beam-search in SAR.

参数

beam_width (int) – Width for beam search.

class mmocr.models.textrecog.decoders.PositionAttentionDecoder(num_classes=None, rnn_layers=2, dim_input=512, dim_model=128, max_seq_len=40, mask=True, return_feature=False, encode_value=False, init_cfg=None)[源代码]
class mmocr.models.textrecog.decoders.RobustScannerDecoder(num_classes=None, dim_input=512, dim_model=128, max_seq_len=40, start_idx=0, mask=True, padding_idx=None, encode_value=False, hybrid_decoder=None, position_decoder=None, init_cfg=None)[源代码]
class mmocr.models.textrecog.decoders.SequenceAttentionDecoder(num_classes=None, rnn_layers=2, dim_input=512, dim_model=128, max_seq_len=40, start_idx=0, mask=True, padding_idx=None, dropout_ratio=0, return_feature=False, encode_value=False, init_cfg=None)[源代码]
class mmocr.models.textrecog.decoders.SequentialSARDecoder(num_classes=37, enc_bi_rnn=False, dec_bi_rnn=False, dec_gru=False, d_k=64, d_model=512, d_enc=512, pred_dropout=0.0, mask=True, max_seq_len=40, start_idx=0, padding_idx=92, pred_concat=False, init_cfg=None, **kwargs)[源代码]

Implementation Sequential Decoder module in `SAR.

<https://arxiv.org/abs/1811.00751>`_.

参数
  • number_classes (int) – Number of output class.

  • enc_bi_rnn (bool) – If True, use bidirectional RNN in encoder.

  • dec_bi_rnn (bool) – If True, use bidirectional RNN in decoder.

  • dec_do_rnn (float) – Dropout of RNN layer in decoder.

  • dec_gru (bool) – If True, use GRU, else LSTM in decoder.

  • d_k (int) – Dim of conv layers in attention module.

  • d_model (int) – Dim of channels from backbone.

  • d_enc (int) – Dim of encoder RNN layer.

  • pred_dropout (float) – Dropout probability of prediction layer.

  • max_seq_len (int) – Maximum sequence length during decoding.

  • mask (bool) – If True, mask padding in feature map.

  • start_idx (int) – Index of start token.

  • padding_idx (int) – Index of padding token.

  • pred_concat (bool) – If True, concat glimpse feature from attention with holistic feature and hidden state.

class mmocr.models.textrecog.decoders.TFDecoder(n_layers=6, d_embedding=512, n_head=8, d_k=64, d_v=64, d_model=512, d_inner=256, n_position=200, dropout=0.1, num_classes=93, max_seq_len=40, start_idx=1, padding_idx=92, init_cfg=None, **kwargs)[源代码]

Transformer Decoder block with self attention mechanism.

textrecog_losses

class mmocr.models.textrecog.losses.CELoss(ignore_index=- 1, reduction='none')[源代码]

Implementation of loss module for encoder-decoder based text recognition method with CrossEntropy loss.

参数
  • ignore_index (int) – Specifies a target value that is ignored and does not contribute to the input gradient.

  • reduction (str) – Specifies the reduction to apply to the output, should be one of the following: (‘none’, ‘mean’, ‘sum’).

forward(outputs, targets_dict, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.losses.CTCLoss(flatten=True, blank=0, reduction='mean', zero_infinity=False, **kwargs)[源代码]

Implementation of loss module for CTC-loss based text recognition.

参数
  • flatten (bool) – If True, use flattened targets, else padded targets.

  • blank (int) – Blank label. Default 0.

  • reduction (str) – Specifies the reduction to apply to the output, should be one of the following: (‘none’, ‘mean’, ‘sum’).

  • zero_infinity (bool) – Whether to zero infinite losses and the associated gradients. Default: False. Infinite losses mainly occur when the inputs are too short to be aligned to the targets.

forward(outputs, targets_dict, img_metas=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.losses.SARLoss(ignore_index=0, reduction='mean', **kwargs)[源代码]

Implementation of loss module in `SAR.

<https://arxiv.org/abs/1811.00751>`_.

参数
  • ignore_index (int) – Specifies a target value that is ignored and does not contribute to the input gradient.

  • reduction (str) – Specifies the reduction to apply to the output, should be one of the following: (‘none’, ‘mean’, ‘sum’).

class mmocr.models.textrecog.losses.SegLoss(seg_downsample_ratio=0.5, seg_with_loss_weight=True, ignore_index=255, **kwargs)[源代码]

Implementation of loss module for segmentation based text recognition method.

参数
  • seg_downsample_ratio (float) – Downsample ratio of segmentation map.

  • seg_with_loss_weight (bool) – If True, set weight for segmentation loss.

  • ignore_index (int) – Specifies a target value that is ignored and does not contribute to the input gradient.

forward(out_neck, out_head, gt_kernels)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.losses.TFLoss(ignore_index=- 1, reduction='none', flatten=True, **kwargs)[源代码]

Implementation of loss module for transformer.

textrecog_backbones

class mmocr.models.textrecog.backbones.NRTRModalityTransform(input_channels=3, input_height=32, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.ResNet31OCR(base_channels=3, layers=[1, 2, 5, 3], channels=[64, 128, 256, 256, 512, 512, 512], out_indices=None, stage4_pool_cfg={'kernel_size': (2, 1), 'stride': (2, 1)}, last_stage_pool=False, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement ResNet backbone for text recognition, modified from

ResNet

参数
  • base_channels (int) – Number of channels of input image tensor.

  • layers (list[int]) – List of BasicBlock number for each stage.

  • channels (list[int]) – List of out_channels of Conv2d layer.

  • out_indices (None | Sequence[int]) – Indices of output stages.

  • stage4_pool_cfg (dict) – Dictionary to construct and configure pooling layer in stage 4.

  • last_stage_pool (bool) – If True, add MaxPool2d layer to last stage.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.ShallowCNN(input_channels=1, hidden_dim=512, init_cfg=[{'type': 'Kaiming', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement Shallow CNN block for SATRN, see

SATRN

参数
  • base_channels (int) – Number of channels of input image tensor.

  • hidden_dim (int) – Size of hidden layers of the model.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.backbones.VeryDeepVgg(leaky_relu=True, input_channels=3, init_cfg=[{'type': 'Xavier', 'layer': 'Conv2d'}, {'type': 'Uniform', 'layer': 'BatchNorm2d'}])[源代码]
Implement VGG-VeryDeep backbone for text recognition, modified from

VGG-VeryDeep

参数
  • leaky_relu (bool) – Use leakyRelu or not.

  • input_channels (int) – Number of channels of input image tensor.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

textrecog_layers

class mmocr.models.textrecog.layers.Adaptive2DPositionalEncoding(d_hid=512, n_height=100, n_width=100, dropout=0.1, init_cfg=[{'type': 'Xavier', 'layer': 'Conv2d'}])[源代码]
Implement Adaptive 2D positional encoder for SATRN, see

SATRN Modified from https://github.com/Media-Smart/vedastr Licensed under the Apache License, Version 2.0 (the “License”);

参数
  • d_hid (int) – Dimensions of hidden layer.

  • n_height (int) – Max height of the 2D feature output.

  • n_width (int) – Max width of the 2D feature output.

  • dropout (int) – Size of hidden layers of the model.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.BasicBlock(inplanes, planes, stride=1, downsample=False)[源代码]
forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.BidirectionalLSTM(nIn, nHidden, nOut)[源代码]
forward(input)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.Bottleneck(inplanes, planes, stride=1, downsample=False)[源代码]
forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.DotProductAttentionLayer(dim_model=None)[源代码]
forward(query, key, value, mask=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.MultiHeadAttention(n_head=8, d_model=512, d_k=64, d_v=64, dropout=0.1, qkv_bias=False, mask_value=0)[源代码]

Multi-Head Attention module.

forward(q, k, v, mask=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.PositionAwareLayer(dim_model, rnn_layers=2)[源代码]
forward(img_feature)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.PositionalEncoding(d_hid=512, n_position=200)[源代码]
forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.PositionwiseFeedForward(d_in, d_hid, dropout=0.1, act_layer=<class 'torch.nn.modules.activation.GELU'>)[源代码]

A two-feed-forward-layer module.

forward(x)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.RobustScannerFusionLayer(dim_model, dim=- 1, init_cfg=None)[源代码]
forward(x0, x1)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class mmocr.models.textrecog.layers.TransformerDecoderLayer(d_model=512, d_inner=256, n_head=8, d_k=64, d_v=64, dropout=0.1, qkv_bias=False, mask_value=0, act_layer=<class 'torch.nn.modules.activation.GELU'>)[源代码]
forward(dec_input, enc_output, self_attn_mask=None, dec_enc_attn_mask=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

mmocr.models.textrecog.layers.get_subsequent_mask(seq)[源代码]

For masking out the subsequent info.

kie_extractors

class mmocr.models.kie.extractors.SDMGR(backbone, neck=None, bbox_head=None, extractor={'featmap_strides': [1], 'roi_layer': {'output_size': 7, 'type': 'RoIAlign'}, 'type': 'mmdet.SingleRoIExtractor'}, visual_modality=False, train_cfg=None, test_cfg=None, class_list=None, init_cfg=None)[源代码]

The implementation of the paper: Spatial Dual-Modality Graph Reasoning for Key Information Extraction. https://arxiv.org/abs/2103.14470.

参数
  • visual_modality (bool) – Whether use the visual modality.

  • class_list (None | str) – Mapping file of class index to class name. If None, class index will be shown in show_results, else class name.

extract_feat(img, gt_bboxes)[源代码]

Directly extract features from the backbone+neck.

forward_test(img, img_metas, relations, texts, gt_bboxes, rescale=False)[源代码]

Args: imgs (List[Tensor]): the outer list indicates test-time

augmentations and inner Tensor should have a shape NxCxHxW, which contains all images in the batch.

img_metas (List[List[dict]]): the outer list indicates test-time

augs (multiscale, flip, etc.) and the inner list indicates images in a batch.

forward_train(img, img_metas, relations, texts, gt_bboxes, gt_labels)[源代码]
参数
  • img (tensor) – Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – A list of image info dict where each dict contains: ‘img_shape’, ‘scale_factor’, ‘flip’, and may also contain ‘filename’, ‘ori_shape’, ‘pad_shape’, and ‘img_norm_cfg’. For details of the values of these keys, please see mmdet.datasets.pipelines.Collect.

  • relations (list[tensor]) – Relations between bboxes.

  • texts (list[tensor]) – Texts in bboxes.

  • gt_bboxes (list[tensor]) – Each item is the truth boxes for each image in [tl_x, tl_y, br_x, br_y] format.

  • gt_labels (list[tensor]) – Class indices corresponding to each box.

返回

A dictionary of loss components.

返回类型

dict[str, tensor]

show_result(img, result, boxes, win_name='', show=False, wait_time=0, out_file=None, **kwargs)[源代码]

Draw result on img.

参数
  • img (str or tensor) – The image to be displayed.

  • result (dict) – The results to draw on img.

  • boxes (list) – Bbox of img.

  • win_name (str) – The window name.

  • wait_time (int) – Value of waitKey param. Default: 0.

  • show (bool) – Whether to show the image. Default: False.

  • out_file (str or None) – The output filename. Default: None.

返回

Only if not show or out_file.

返回类型

img (tensor)

kie_heads

class mmocr.models.kie.heads.SDMGRHead(num_chars=92, visual_dim=64, fusion_dim=1024, node_input=32, node_embed=256, edge_input=5, edge_embed=256, num_gnn=2, num_classes=26, loss={'type': 'SDMGRLoss'}, bidirectional=False, train_cfg=None, test_cfg=None, init_cfg={'mean': 0, 'override': {'name': 'edge_embed'}, 'std': 0.01, 'type': 'Normal'})[源代码]
forward(relations, texts, x=None)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

kie_losses

class mmocr.models.kie.losses.SDMGRLoss(node_weight=1.0, edge_weight=1.0, ignore=- 100)[源代码]

The implementation the loss of key information extraction proposed in the paper: Spatial Dual-Modality Graph Reasoning for Key Information Extraction.

https://arxiv.org/abs/2103.14470.

forward(node_preds, edge_preds, gts)[源代码]

Defines the computation performed at every call.

Should be overridden by all subclasses.

注解

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

mmocr.datasets

class mmocr.datasets.BaseDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]

Custom dataset for text detection, text recognition, and their downstream tasks.

  1. The text detection annotation format is as follows: The annotations field is optional for testing (this is one line of anno_file, with line-json-str

    converted to dict for visualizing only).

    {

    “file_name”: “sample.jpg”, “height”: 1080, “width”: 960, “annotations”:

    [
    {

    “iscrowd”: 0, “category_id”: 1, “bbox”: [357.0, 667.0, 804.0, 100.0], “segmentation”: [[361, 667, 710, 670,

    72, 767, 357, 763]]

    }

    ]

    }

  2. The two text recognition annotation formats are as follows: The x1,y1,x2,y2,x3,y3,x4,y4 field is used for online crop augmentation during training.

    format1: sample.jpg hello format2: sample.jpg 20 20 100 20 100 40 20 40 hello

参数
  • ann_file (str) – Annotation file path.

  • pipeline (list[dict]) – Processing pipeline.

  • loader (dict) – Dictionary to construct loader to load annotation infos.

  • img_prefix (str, optional) – Image prefix to generate full image path.

  • test_mode (bool, optional) – If set True, try…except will be turned off in __getitem__.

evaluate(results, metric=None, logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

format_results(results, **kwargs)[源代码]

Placeholder to format result to dataset-specific output.

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_test_img(img_info)[源代码]

Get testing data from pipeline.

参数

idx (int) – Index of data.

返回

Testing data after pipeline with new keys introduced by

pipeline.

返回类型

dict

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.CustomFormatBundle(keys=[], call_super=True, visualize={'boundary_key': None, 'flag': False})[源代码]

Custom formatting bundle.

It formats common fields such as ‘img’ and ‘proposals’ as done in DefaultFormatBundle, while other fields such as ‘gt_kernels’ and ‘gt_effective_region_mask’ will be formatted to DC as follows:

  • gt_kernels: to DataContainer (cpu_only=True)

  • gt_effective_mask: to DataContainer (cpu_only=True)

参数
  • keys (list[str]) – Fields to be formatted to DC only.

  • call_super (bool) – If True, format common fields by DefaultFormatBundle, else format fields in keys above only.

  • visualize (dict) – If flag=True, visualize gt mask for debugging.

class mmocr.datasets.DBNetTargets(shrink_ratio=0.4, thr_min=0.3, thr_max=0.7, min_short_size=8)[源代码]

Generate gt shrunk text, gt threshold map, and their effective region masks to learn DBNet: Real-time Scene Text Detection with Differentiable Binarization [https://arxiv.org/abs/1911.08947]. This was partially adapted from https://github.com/MhLiao/DB.

参数
  • shrink_ratio (float) – The area shrunk ratio between text kernels and their text masks.

  • thr_min (float) – The minimum value of the threshold map.

  • thr_max (float) – The maximum value of the threshold map.

  • min_short_size (int) – The minimum size of polygon below which the polygon is invalid.

draw_border_map(polygon, canvas, mask)[源代码]

Generate threshold map for one polygon.

参数
  • polygon (ndarray) – The polygon boundary ndarray.

  • canvas (ndarray) – The generated threshold map.

  • mask (ndarray) – The generated threshold mask.

find_invalid(results)[源代码]

Find invalid polygons.

参数

results (dict) – The dict containing gt_mask.

返回

The indicators for ignoring polygons.

返回类型

ignore_tags (list[bool])

generate_targets(results)[源代码]

Generate the gt targets for DBNet.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

generate_thr_map(img_size, polygons)[源代码]

Generate threshold map.

参数
  • img_size (tuple(int)) – The image size (h,w)

  • polygons (list(ndarray)) – The polygon list.

返回

The generated threshold map. thr_mask (ndarray): The effective mask of threshold map.

返回类型

thr_map (ndarray)

ignore_texts(results, ignore_tags)[源代码]

Ignore gt masks and gt_labels while padding gt_masks_ignore in results given ignore_tags.

参数
  • results (dict) – Result for one image.

  • ignore_tags (list[int]) – Indicate whether to ignore its corresponding ground truth text.

返回

Results after filtering.

返回类型

results (dict)

invalid_polygon(poly)[源代码]

Judge the input polygon is invalid or not. It is invalid if its area smaller than 1 or the shorter side of its minimum bounding box smaller than min_short_size.

参数

poly (ndarray) – The polygon boundary point sequence.

返回

Whether the polygon is invalid.

返回类型

True/False (bool)

class mmocr.datasets.FCENetTargets(fourier_degree=5, resample_step=4.0, center_region_shrink_ratio=0.3, level_size_divisors=(8, 16, 32), level_proportion_range=((0, 0.4), (0.3, 0.7), (0.6, 1.0)))[源代码]

Generate the ground truth targets of FCENet: Fourier Contour Embedding for Arbitrary-Shaped Text Detection.

[https://arxiv.org/abs/2104.10442]

参数
  • fourier_degree (int) – The maximum Fourier transform degree k.

  • resample_step (float) – The step size for resampling the text center line (TCL). It’s better not to exceed half of the minimum width.

  • center_region_shrink_ratio (float) – The shrink ratio of text center region.

  • level_size_divisors (tuple(int)) – The downsample ratio on each level.

  • level_proportion_range (tuple(tuple(int))) – The range of text sizes assigned to each level.

cal_fourier_signature(polygon, fourier_degree)[源代码]

Calculate Fourier signature from input polygon.

参数
  • polygon (ndarray) – The input polygon.

  • fourier_degree (int) – The maximum Fourier degree K.

返回

An array shaped (2k+1, 2) containing

real part and image part of 2k+1 Fourier coefficients.

返回类型

fourier_signature (ndarray)

clockwise(c, fourier_degree)[源代码]

Make sure the polygon reconstructed from Fourier coefficients c in the clockwise direction.

参数

polygon (list[float]) – The origin polygon.

返回

The polygon in clockwise point order.

返回类型

new_polygon (lost[float])

generate_center_region_mask(img_size, text_polys)[源代码]

Generate text center region mask.

参数
  • img_size (tuple) – The image size of (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The text center region mask.

返回类型

center_region_mask (ndarray)

generate_fourier_maps(img_size, text_polys)[源代码]

Generate Fourier coefficient maps.

参数
  • img_size (tuple) – The image size of (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The Fourier coefficient real part maps. fourier_image_map (ndarray): The Fourier coefficient image part

maps.

返回类型

fourier_real_map (ndarray)

generate_level_targets(img_size, text_polys, ignore_polys)[源代码]

Generate ground truth target on each level.

参数
  • img_size (list[int]) – Shape of input image.

  • text_polys (list[list[ndarray]]) – A list of ground truth polygons.

  • ignore_polys (list[list[ndarray]]) – A list of ignored polygons.

返回

A list of ground target on each level.

返回类型

level_maps (list(ndarray))

generate_targets(results)[源代码]

Generate the ground truth targets for FCENet.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

normalize_polygon(polygon)[源代码]

Normalize one polygon so that its start point is at right most.

参数

polygon (list[float]) – The origin polygon.

返回

The polygon with start point at right.

返回类型

new_polygon (lost[float])

poly2fourier(polygon, fourier_degree)[源代码]

Perform Fourier transformation to generate Fourier coefficients ck from polygon.

参数
  • polygon (ndarray) – An input polygon.

  • fourier_degree (int) – The maximum Fourier degree K.

返回

Fourier coefficients.

返回类型

c (ndarray(complex))

resample_polygon(polygon, n=400)[源代码]

Resample one polygon with n points on its boundary.

参数
  • polygon (list[float]) – The input polygon.

  • n (int) – The number of resampled points.

返回

The resampled polygon.

返回类型

resampled_polygon (list[float])

class mmocr.datasets.HardDiskLoader(ann_file, parser, repeat=1)[源代码]

Load annotation file from hard disk to RAM.

参数

ann_file (str) – Annotation file path.

class mmocr.datasets.IcdarDataset(ann_file, pipeline, classes=None, data_root=None, img_prefix='', seg_prefix=None, proposal_file=None, test_mode=False, filter_empty_gt=True, select_first_k=- 1)[源代码]
evaluate(results, metric='hmean-iou', logger=None, score_thr=0.3, rank_list=None, **kwargs)[源代码]

Evaluate the hmean metric.

参数
  • results (list[dict]) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

  • rank_list (str) – json file used to save eval result of each image after ranking.

返回

float]]: The evaluation results.

返回类型

dict[dict[str

load_annotations(ann_file)[源代码]

Load annotation from COCO style annotation file.

参数

ann_file (str) – Path of annotation file.

返回

Annotation info from COCO api.

返回类型

list[dict]

class mmocr.datasets.KIEDataset(ann_file=None, loader=None, dict_file=None, img_prefix='', pipeline=None, norm=10.0, directed=False, test_mode=True, **kwargs)[源代码]
参数
  • ann_file (str) – Annotation file path.

  • pipeline (list[dict]) – Processing pipeline.

  • loader (dict) – Dictionary to construct loader to load annotation infos.

  • img_prefix (str, optional) – Image prefix to generate full image path.

  • test_mode (bool, optional) – If True, try…except will be turned off in __getitem__.

  • dict_file (str) – Character dict file path.

  • norm (float) – Norm to map value from one range to another.

compute_relation(boxes)[源代码]

Compute relation between every two boxes.

evaluate(results, metric='macro_f1', metric_options={'macro_f1': {'ignores': []}}, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

list_to_numpy(ann_infos)[源代码]

Convert bboxes, relations, texts and labels to ndarray.

pad_text_indices(text_inds)[源代码]

Pad text index to same length.

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.LineJsonParser(keys=[])[源代码]

Parse json-string of one line in annotation file to dict format.

参数

keys (list[str]) – Keys in both json-string and result dict.

class mmocr.datasets.LineStrParser(keys=['filename', 'text'], keys_idx=[0, 1], separator=' ', **kwargs)[源代码]

Parse string of one line in annotation file to dict format.

参数
  • keys (list[str]) – Keys in result dict.

  • keys_idx (list[int]) – Value index in sub-string list for each key above.

  • separator (str) – Separator to separate string to list of sub-string.

class mmocr.datasets.LmdbLoader(ann_file, parser, repeat=1)[源代码]

Load annotation file with lmdb storage backend.

class mmocr.datasets.NerDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]

Custom dataset for named entity recognition tasks.

参数
  • ann_file (txt) – Annotation file path.

  • loader (dict) – Dictionary to construct loader to load annotation infos.

  • pipeline (list[dict]) – Processing pipeline.

  • test_mode (bool, optional) – If True, try…except will be turned off in __getitem__.

evaluate(results, metric=None, logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

A dict containing the following keys:

’acc’, ‘recall’, ‘f1-score’.

返回类型

info (dict)

prepare_train_img(index)[源代码]

Get training data and annotations after pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys introduced by pipeline.

返回类型

dict

class mmocr.datasets.OCRDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
evaluate(results, metric='acc', logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

class mmocr.datasets.OCRSegDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.TextDetDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
evaluate(results, metric='hmean-iou', score_thr=0.3, rank_list=None, logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • score_thr (float) – Score threshold for prediction map.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

  • rank_list (str) – json file used to save eval result of each image after ranking.

返回

float]

返回类型

dict[str

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.UniformConcatDataset(datasets, separate_eval=True, pipeline=None, **kwargs)[源代码]

A wrapper of concatenated dataset.

Same as torch.utils.data.dataset.ConcatDataset, but concat the group flag for image aspect ratio.

参数
  • datasets (list[Dataset]) – A list of datasets.

  • separate_eval (bool) – Whether to evaluate the results separately if it is used as validation dataset. Defaults to True.

mmocr.datasets.build_dataloader(dataset, samples_per_gpu, workers_per_gpu, num_gpus=1, dist=True, shuffle=True, seed=None, **kwargs)[源代码]

Build PyTorch DataLoader.

In distributed training, each GPU/process has a dataloader. In non-distributed training, there is only one dataloader for all GPUs.

参数
  • dataset (Dataset) – A PyTorch dataset.

  • samples_per_gpu (int) – Number of training samples on each GPU, i.e., batch size of each GPU.

  • workers_per_gpu (int) – How many subprocesses to use for data loading for each GPU.

  • num_gpus (int) – Number of GPUs. Only used in non-distributed training.

  • dist (bool) – Distributed training/test or not. Default: True.

  • shuffle (bool) – Whether to shuffle the data at every epoch. Default: True.

  • kwargs – any keyword argument to be used to initialize DataLoader

返回

A PyTorch dataloader.

返回类型

DataLoader

datasets

class mmocr.datasets.base_dataset.BaseDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]

Custom dataset for text detection, text recognition, and their downstream tasks.

  1. The text detection annotation format is as follows: The annotations field is optional for testing (this is one line of anno_file, with line-json-str

    converted to dict for visualizing only).

    {

    “file_name”: “sample.jpg”, “height”: 1080, “width”: 960, “annotations”:

    [
    {

    “iscrowd”: 0, “category_id”: 1, “bbox”: [357.0, 667.0, 804.0, 100.0], “segmentation”: [[361, 667, 710, 670,

    72, 767, 357, 763]]

    }

    ]

    }

  2. The two text recognition annotation formats are as follows: The x1,y1,x2,y2,x3,y3,x4,y4 field is used for online crop augmentation during training.

    format1: sample.jpg hello format2: sample.jpg 20 20 100 20 100 40 20 40 hello

参数
  • ann_file (str) – Annotation file path.

  • pipeline (list[dict]) – Processing pipeline.

  • loader (dict) – Dictionary to construct loader to load annotation infos.

  • img_prefix (str, optional) – Image prefix to generate full image path.

  • test_mode (bool, optional) – If set True, try…except will be turned off in __getitem__.

evaluate(results, metric=None, logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

format_results(results, **kwargs)[源代码]

Placeholder to format result to dataset-specific output.

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_test_img(img_info)[源代码]

Get testing data from pipeline.

参数

idx (int) – Index of data.

返回

Testing data after pipeline with new keys introduced by

pipeline.

返回类型

dict

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.icdar_dataset.IcdarDataset(ann_file, pipeline, classes=None, data_root=None, img_prefix='', seg_prefix=None, proposal_file=None, test_mode=False, filter_empty_gt=True, select_first_k=- 1)[源代码]
evaluate(results, metric='hmean-iou', logger=None, score_thr=0.3, rank_list=None, **kwargs)[源代码]

Evaluate the hmean metric.

参数
  • results (list[dict]) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

  • rank_list (str) – json file used to save eval result of each image after ranking.

返回

float]]: The evaluation results.

返回类型

dict[dict[str

load_annotations(ann_file)[源代码]

Load annotation from COCO style annotation file.

参数

ann_file (str) – Path of annotation file.

返回

Annotation info from COCO api.

返回类型

list[dict]

class mmocr.datasets.ocr_dataset.OCRDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
evaluate(results, metric='acc', logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

class mmocr.datasets.ocr_seg_dataset.OCRSegDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.text_det_dataset.TextDetDataset(ann_file, loader, pipeline, img_prefix='', test_mode=False)[源代码]
evaluate(results, metric='hmean-iou', score_thr=0.3, rank_list=None, logger=None, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • score_thr (float) – Score threshold for prediction map.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

  • rank_list (str) – json file used to save eval result of each image after ranking.

返回

float]

返回类型

dict[str

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

class mmocr.datasets.kie_dataset.KIEDataset(ann_file=None, loader=None, dict_file=None, img_prefix='', pipeline=None, norm=10.0, directed=False, test_mode=True, **kwargs)[源代码]
参数
  • ann_file (str) – Annotation file path.

  • pipeline (list[dict]) – Processing pipeline.

  • loader (dict) – Dictionary to construct loader to load annotation infos.

  • img_prefix (str, optional) – Image prefix to generate full image path.

  • test_mode (bool, optional) – If True, try…except will be turned off in __getitem__.

  • dict_file (str) – Character dict file path.

  • norm (float) – Norm to map value from one range to another.

compute_relation(boxes)[源代码]

Compute relation between every two boxes.

evaluate(results, metric='macro_f1', metric_options={'macro_f1': {'ignores': []}}, **kwargs)[源代码]

Evaluate the dataset.

参数
  • results (list) – Testing results of the dataset.

  • metric (str | list[str]) – Metrics to be evaluated.

  • logger (logging.Logger | str | None) – Logger used for printing related information during evaluation. Default: None.

返回

float]

返回类型

dict[str

list_to_numpy(ann_infos)[源代码]

Convert bboxes, relations, texts and labels to ndarray.

pad_text_indices(text_inds)[源代码]

Pad text index to same length.

pre_pipeline(results)[源代码]

Prepare results dict for pipeline.

prepare_train_img(index)[源代码]

Get training data and annotations from pipeline.

参数

index (int) – Index of data.

返回

Training data and annotation after pipeline with new keys

introduced by pipeline.

返回类型

dict

pipelines

class mmocr.datasets.pipelines.ColorJitter(**kwargs)[源代码]

An interface for torch color jitter so that it can be invoked in mmdetection pipeline.

class mmocr.datasets.pipelines.CustomFormatBundle(keys=[], call_super=True, visualize={'boundary_key': None, 'flag': False})[源代码]

Custom formatting bundle.

It formats common fields such as ‘img’ and ‘proposals’ as done in DefaultFormatBundle, while other fields such as ‘gt_kernels’ and ‘gt_effective_region_mask’ will be formatted to DC as follows:

  • gt_kernels: to DataContainer (cpu_only=True)

  • gt_effective_mask: to DataContainer (cpu_only=True)

参数
  • keys (list[str]) – Fields to be formatted to DC only.

  • call_super (bool) – If True, format common fields by DefaultFormatBundle, else format fields in keys above only.

  • visualize (dict) – If flag=True, visualize gt mask for debugging.

class mmocr.datasets.pipelines.DBNetTargets(shrink_ratio=0.4, thr_min=0.3, thr_max=0.7, min_short_size=8)[源代码]

Generate gt shrunk text, gt threshold map, and their effective region masks to learn DBNet: Real-time Scene Text Detection with Differentiable Binarization [https://arxiv.org/abs/1911.08947]. This was partially adapted from https://github.com/MhLiao/DB.

参数
  • shrink_ratio (float) – The area shrunk ratio between text kernels and their text masks.

  • thr_min (float) – The minimum value of the threshold map.

  • thr_max (float) – The maximum value of the threshold map.

  • min_short_size (int) – The minimum size of polygon below which the polygon is invalid.

draw_border_map(polygon, canvas, mask)[源代码]

Generate threshold map for one polygon.

参数
  • polygon (ndarray) – The polygon boundary ndarray.

  • canvas (ndarray) – The generated threshold map.

  • mask (ndarray) – The generated threshold mask.

find_invalid(results)[源代码]

Find invalid polygons.

参数

results (dict) – The dict containing gt_mask.

返回

The indicators for ignoring polygons.

返回类型

ignore_tags (list[bool])

generate_targets(results)[源代码]

Generate the gt targets for DBNet.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

generate_thr_map(img_size, polygons)[源代码]

Generate threshold map.

参数
  • img_size (tuple(int)) – The image size (h,w)

  • polygons (list(ndarray)) – The polygon list.

返回

The generated threshold map. thr_mask (ndarray): The effective mask of threshold map.

返回类型

thr_map (ndarray)

ignore_texts(results, ignore_tags)[源代码]

Ignore gt masks and gt_labels while padding gt_masks_ignore in results given ignore_tags.

参数
  • results (dict) – Result for one image.

  • ignore_tags (list[int]) – Indicate whether to ignore its corresponding ground truth text.

返回

Results after filtering.

返回类型

results (dict)

invalid_polygon(poly)[源代码]

Judge the input polygon is invalid or not. It is invalid if its area smaller than 1 or the shorter side of its minimum bounding box smaller than min_short_size.

参数

poly (ndarray) – The polygon boundary point sequence.

返回

Whether the polygon is invalid.

返回类型

True/False (bool)

class mmocr.datasets.pipelines.FCENetTargets(fourier_degree=5, resample_step=4.0, center_region_shrink_ratio=0.3, level_size_divisors=(8, 16, 32), level_proportion_range=((0, 0.4), (0.3, 0.7), (0.6, 1.0)))[源代码]

Generate the ground truth targets of FCENet: Fourier Contour Embedding for Arbitrary-Shaped Text Detection.

[https://arxiv.org/abs/2104.10442]

参数
  • fourier_degree (int) – The maximum Fourier transform degree k.

  • resample_step (float) – The step size for resampling the text center line (TCL). It’s better not to exceed half of the minimum width.

  • center_region_shrink_ratio (float) – The shrink ratio of text center region.

  • level_size_divisors (tuple(int)) – The downsample ratio on each level.

  • level_proportion_range (tuple(tuple(int))) – The range of text sizes assigned to each level.

cal_fourier_signature(polygon, fourier_degree)[源代码]

Calculate Fourier signature from input polygon.

参数
  • polygon (ndarray) – The input polygon.

  • fourier_degree (int) – The maximum Fourier degree K.

返回

An array shaped (2k+1, 2) containing

real part and image part of 2k+1 Fourier coefficients.

返回类型

fourier_signature (ndarray)

clockwise(c, fourier_degree)[源代码]

Make sure the polygon reconstructed from Fourier coefficients c in the clockwise direction.

参数

polygon (list[float]) – The origin polygon.

返回

The polygon in clockwise point order.

返回类型

new_polygon (lost[float])

generate_center_region_mask(img_size, text_polys)[源代码]

Generate text center region mask.

参数
  • img_size (tuple) – The image size of (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The text center region mask.

返回类型

center_region_mask (ndarray)

generate_fourier_maps(img_size, text_polys)[源代码]

Generate Fourier coefficient maps.

参数
  • img_size (tuple) – The image size of (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The Fourier coefficient real part maps. fourier_image_map (ndarray): The Fourier coefficient image part

maps.

返回类型

fourier_real_map (ndarray)

generate_level_targets(img_size, text_polys, ignore_polys)[源代码]

Generate ground truth target on each level.

参数
  • img_size (list[int]) – Shape of input image.

  • text_polys (list[list[ndarray]]) – A list of ground truth polygons.

  • ignore_polys (list[list[ndarray]]) – A list of ignored polygons.

返回

A list of ground target on each level.

返回类型

level_maps (list(ndarray))

generate_targets(results)[源代码]

Generate the ground truth targets for FCENet.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

normalize_polygon(polygon)[源代码]

Normalize one polygon so that its start point is at right most.

参数

polygon (list[float]) – The origin polygon.

返回

The polygon with start point at right.

返回类型

new_polygon (lost[float])

poly2fourier(polygon, fourier_degree)[源代码]

Perform Fourier transformation to generate Fourier coefficients ck from polygon.

参数
  • polygon (ndarray) – An input polygon.

  • fourier_degree (int) – The maximum Fourier degree K.

返回

Fourier coefficients.

返回类型

c (ndarray(complex))

resample_polygon(polygon, n=400)[源代码]

Resample one polygon with n points on its boundary.

参数
  • polygon (list[float]) – The input polygon.

  • n (int) – The number of resampled points.

返回

The resampled polygon.

返回类型

resampled_polygon (list[float])

class mmocr.datasets.pipelines.FancyPCA(eig_vec=None, eig_val=None)[源代码]

Implementation of PCA based image augmentation, proposed in the paper Imagenet Classification With Deep Convolutional Neural Networks.

It alters the intensities of RGB values along the principal components of ImageNet dataset.

class mmocr.datasets.pipelines.ImgAug(args=None)[源代码]

A wrapper to use imgaug https://github.com/aleju/imgaug.

参数

args ([list[list|dict]]) – The argumentation list. For details, please refer to imgaug document. Take args=[[‘Fliplr’, 0.5], dict(cls=’Affine’, rotate=[-10, 10]), [‘Resize’, [0.5, 3.0]]] as an example. The args horizontally flip images with probability 0.5, followed by random rotation with angles in range [-10, 10], and resize with an independent scale in range [0.5, 3.0] for each side of images.

class mmocr.datasets.pipelines.KIEFormatBundle[源代码]

Key information extraction formatting bundle.

Based on the DefaultFormatBundle, itt simplifies the pipeline of formatting common fields, including “img”, “proposals”, “gt_bboxes”, “gt_labels”, “gt_masks”, “gt_semantic_seg”, “relations” and “texts”. These fields are formatted as follows.

  • img: (1) transpose, (2) to tensor, (3) to DataContainer (stack=True)

  • proposals: (1) to tensor, (2) to DataContainer

  • gt_bboxes: (1) to tensor, (2) to DataContainer

  • gt_bboxes_ignore: (1) to tensor, (2) to DataContainer

  • gt_labels: (1) to tensor, (2) to DataContainer

  • gt_masks: (1) to tensor, (2) to DataContainer (cpu_only=True)

  • gt_semantic_seg: (1) unsqueeze dim-0 (2) to tensor,
    1. to DataContainer (stack=True)

  • relations: (1) scale, (2) to tensor, (3) to DataContainer

  • texts: (1) to tensor, (2) to DataContainer

class mmocr.datasets.pipelines.LoadImageFromNdarray(to_float32=False, color_type='color', file_client_args={'backend': 'disk'})[源代码]

Load an image from np.ndarray.

Similar with LoadImageFromFile, but the image read from results['img'], which is np.ndarray.

class mmocr.datasets.pipelines.LoadTextAnnotations(with_bbox=True, with_label=True, with_mask=False, with_seg=False, poly2mask=True)[源代码]
process_polygons(polygons)[源代码]

Convert polygons to list of ndarray and filter invalid polygons.

参数

polygons (list[list]) – Polygons of one instance.

返回

Processed polygons.

返回类型

list[numpy.ndarray]

class mmocr.datasets.pipelines.MultiRotateAugOCR(transforms, rotate_degrees=None, force_rotate=False)[源代码]

Test-time augmentation with multiple rotations in the case that img_height > img_width.

An example configuration is as follows:

rotate_degrees=[0, 90, 270],
transforms=[
    dict(
        type='ResizeOCR',
        height=32,
        min_width=32,
        max_width=160,
        keep_aspect_ratio=True),
    dict(type='ToTensorOCR'),
    dict(type='NormalizeOCR', **img_norm_cfg),
    dict(
        type='Collect',
        keys=['img'],
        meta_keys=[
            'filename', 'ori_shape', 'img_shape', 'valid_ratio'
        ]),
]

After MultiRotateAugOCR with above configuration, the results are wrapped into lists of the same length as follows:

dict(
    img=[...],
    img_shape=[...]
    ...
)
参数
  • transforms (list[dict]) – Transformation applied for each augmentation.

  • rotate_degrees (list[int] | None) – Degrees of anti-clockwise rotation.

  • force_rotate (bool) – If True, rotate image by ‘rotate_degrees’ while ignore image aspect ratio.

class mmocr.datasets.pipelines.NerTransform(label_convertor, max_len)[源代码]

Convert text to ID and entity in ground truth to label ID. The masks and tokens are generated at the same time. The four parameters will be used as input to the model.

参数
  • label_convertor – Convert text to ID and entity

  • ground truth to label ID. (in) –

  • max_len (int) – Limited maximum input length.

class mmocr.datasets.pipelines.NormalizeOCR(mean, std)[源代码]

Normalize a tensor image with mean and standard deviation.

class mmocr.datasets.pipelines.OCRSegTargets(label_convertor=None, attn_shrink_ratio=0.5, seg_shrink_ratio=0.25, box_type='char_rects', pad_val=255)[源代码]

Generate gt shrunk kernels for segmentation based OCR framework.

参数
  • label_convertor (dict) – Dictionary to construct label_convertor to convert char to index.

  • attn_shrink_ratio (float) – The area shrunk ratio between attention kernels and gt text masks.

  • seg_shrink_ratio (float) – The area shrunk ratio between segmentation kernels and gt text masks.

  • box_type (str) – Character box type, should be either ‘char_rects’ or ‘char_quads’, with ‘char_rects’ for rectangle with xyxy style and ‘char_quads’ for quadrangle with x1y1x2y2x3y3x4y4 style.

generate_kernels(resize_shape, pad_shape, char_boxes, char_inds, shrink_ratio=0.5, binary=True)[源代码]

Generate char instance kernels for one shrink ratio.

参数
  • resize_shape (tuple(int, int)) – Image size (height, width) after resizing.

  • pad_shape (tuple(int, int)) – Image size (height, width) after padding.

  • char_boxes (list[list[float]]) – The list of char polygons.

  • char_inds (list[int]) – List of char indexes.

  • shrink_ratio (float) – The shrink ratio of kernel.

  • binary (bool) – If True, return binary ndarray containing 0 & 1 only.

返回

The text kernel mask of (height, width).

返回类型

char_kernel (ndarray)

shrink_char_quad(char_quad, shrink_ratio)[源代码]

Shrink char box in style of quadrangle.

参数
  • char_quad (list[float]) – Char box with format [x1, y1, x2, y2, x3, y3, x4, y4].

  • shrink_ratio (float) – The area shrunk ratio between gt kernels and gt text masks.

shrink_char_rect(char_rect, shrink_ratio)[源代码]

Shrink char box in style of rectangle.

参数
  • char_rect (list[float]) – Char box with format [x_min, y_min, x_max, y_max].

  • shrink_ratio (float) – The area shrunk ratio between gt kernels and gt text masks.

class mmocr.datasets.pipelines.OnlineCropOCR(box_keys=['x1', 'y1', 'x2', 'y2', 'x3', 'y3', 'x4', 'y4'], jitter_prob=0.5, max_jitter_ratio_x=0.05, max_jitter_ratio_y=0.02)[源代码]

Crop text areas from whole image with bounding box jitter. If no bbox is given, return directly.

参数
  • box_keys (list[str]) – Keys in results which correspond to RoI bbox.

  • jitter_prob (float) – The probability of box jitter.

  • max_jitter_ratio_x (float) – Maximum horizontal jitter ratio relative to height.

  • max_jitter_ratio_y (float) – Maximum vertical jitter ratio relative to height.

class mmocr.datasets.pipelines.OpencvToPil(**kwargs)[源代码]

Convert numpy.ndarray (bgr) to PIL Image (rgb).

class mmocr.datasets.pipelines.PANetTargets(shrink_ratio=(1.0, 0.5), max_shrink=20)[源代码]

Generate the ground truths for PANet: Efficient and Accurate Arbitrary- Shaped Text Detection with Pixel Aggregation Network.

[https://arxiv.org/abs/1908.05900]. This code is partially adapted from https://github.com/WenmuZhou/PAN.pytorch.

参数
  • shrink_ratio (tuple[float]) – The ratios for shrinking text instances.

  • max_shrink (int) – The maximum shrink distance.

generate_targets(results)[源代码]

Generate the gt targets for PANet.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

class mmocr.datasets.pipelines.PilToOpencv(**kwargs)[源代码]

Convert PIL Image (rgb) to numpy.ndarray (bgr).

class mmocr.datasets.pipelines.RandomCropInstances(target_size, instance_key, mask_type='inx0', positive_sample_ratio=0.625)[源代码]

Randomly crop images and make sure to contain text instances.

参数
  • target_size (tuple or int) – (height, width)

  • positive_sample_ratio (float) – The probability of sampling regions that go through positive regions.

class mmocr.datasets.pipelines.RandomCropPolyInstances(instance_key='gt_masks', crop_ratio=0.625, min_side_ratio=0.4)[源代码]

Randomly crop images and make sure to contain at least one intact instance.

sample_crop_box(img_size, results)[源代码]

Generate crop box and make sure not to crop the polygon instances.

参数
  • img_size (tuple(int)) – The image size (h, w).

  • results (dict) – The results dict.

class mmocr.datasets.pipelines.RandomPaddingOCR(max_ratio=None, box_type=None)[源代码]

Pad the given image on all sides, as well as modify the coordinates of character bounding box in image.

参数
  • max_ratio (list[int]) – [left, top, right, bottom].

  • box_type (None|str) – Character box type. If not none, should be either ‘char_rects’ or ‘char_quads’, with ‘char_rects’ for rectangle with xyxy style and ‘char_quads’ for quadrangle with x1y1x2y2x3y3x4y4 style.

class mmocr.datasets.pipelines.RandomRotateImageBox(min_angle=- 10, max_angle=10, box_type='char_quads')[源代码]

Rotate augmentation for segmentation based text recognition.

参数
  • min_angle (int) – Minimum rotation angle for image and box.

  • max_angle (int) – Maximum rotation angle for image and box.

  • box_type (str) – Character box type, should be either ‘char_rects’ or ‘char_quads’, with ‘char_rects’ for rectangle with xyxy style and ‘char_quads’ for quadrangle with x1y1x2y2x3y3x4y4 style.

class mmocr.datasets.pipelines.RandomRotateTextDet(rotate_ratio=1.0, max_angle=10)[源代码]

Randomly rotate images.

class mmocr.datasets.pipelines.ResizeNoImg(img_scale, keep_ratio=True)[源代码]

Image resizing without img.

Used for KIE.

class mmocr.datasets.pipelines.ResizeOCR(height, min_width=None, max_width=None, keep_aspect_ratio=True, img_pad_value=0, width_downsample_ratio=0.0625, backend=None)[源代码]

Image resizing and padding for OCR.

参数
  • height (int | tuple(int)) – Image height after resizing.

  • min_width (none | int | tuple(int)) – Image minimum width after resizing.

  • max_width (none | int | tuple(int)) – Image maximum width after resizing.

  • keep_aspect_ratio (bool) – Keep image aspect ratio if True during resizing, Otherwise resize to the size height * max_width.

  • img_pad_value (int) – Scalar to fill padding area.

  • width_downsample_ratio (float) – Downsample ratio in horizontal direction from input image to output feature.

  • backend (str | None) – The image resize backend type. Options are cv2, pillow, None. If backend is None, the global imread_backend specified by mmcv.use_backend() will be used. Default: None.

class mmocr.datasets.pipelines.ScaleAspectJitter(img_scale=None, multiscale_mode='range', ratio_range=None, keep_ratio=False, resize_type='around_min_img_scale', aspect_ratio_range=None, long_size_bound=None, short_size_bound=None, scale_range=None)[源代码]

Resize image and segmentation mask encoded by coordinates.

Allowed resize types are around_min_img_scale, long_short_bound, and indep_sample_in_range.

class mmocr.datasets.pipelines.TextSnakeTargets(orientation_thr=2.0, resample_step=4.0, center_region_shrink_ratio=0.3)[源代码]

Generate the ground truth targets of TextSnake: TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes.

[https://arxiv.org/abs/1807.01544]. This was partially adapted from https://github.com/princewang1994/TextSnake.pytorch.

参数

orientation_thr (float) – The threshold for distinguishing between head edge and tail edge among the horizontal and vertical edges of a quadrangle.

draw_center_region_maps(top_line, bot_line, center_line, center_region_mask, radius_map, sin_map, cos_map, region_shrink_ratio)[源代码]

Draw attributes on text center region.

参数
  • top_line (ndarray) – The points composing top curved sideline of text polygon.

  • bot_line (ndarray) – The points composing bottom curved sideline of text polygon.

  • center_line (ndarray) – The points composing the center line of text instance.

  • center_region_mask (ndarray) – The text center region mask.

  • radius_map (ndarray) – The map where the distance from point to sidelines will be drawn on for each pixel in text center region.

  • sin_map (ndarray) – The map where vector_sin(theta) will be drawn on text center regions. Theta is the angle between tangent line and vector (1, 0).

  • cos_map (ndarray) – The map where vector_cos(theta) will be drawn on text center regions. Theta is the angle between tangent line and vector (1, 0).

  • region_shrink_ratio (float) – The shrink ratio of text center.

find_head_tail(points, orientation_thr)[源代码]

Find the head edge and tail edge of a text polygon.

参数
  • points (ndarray) – The points composing a text polygon.

  • orientation_thr (float) – The threshold for distinguishing between head edge and tail edge among the horizontal and vertical edges of a quadrangle.

返回

The indexes of two points composing head edge. tail_inds (list): The indexes of two points composing tail edge.

返回类型

head_inds (list)

generate_center_mask_attrib_maps(img_size, text_polys)[源代码]

Generate text center region mask and geometric attribute maps.

参数
  • img_size (tuple) – The image size of (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The text center region mask. radius_map (ndarray): The distance map from each pixel in text

center region to top sideline.

sin_map (ndarray): The sin(theta) map where theta is the angle

between vector (top point - bottom point) and vector (1, 0).

cos_map (ndarray): The cos(theta) map where theta is the angle

between vector (top point - bottom point) and vector (1, 0).

返回类型

center_region_mask (ndarray)

generate_targets(results)[源代码]

Generate the gt targets for TextSnake.

参数

results (dict) – The input result dictionary.

返回

The output result dictionary.

返回类型

results (dict)

generate_text_region_mask(img_size, text_polys)[源代码]

Generate text center region mask and geometry attribute maps.

参数
  • img_size (tuple) – The image size (height, width).

  • text_polys (list[list[ndarray]]) – The list of text polygons.

返回

The text region mask.

返回类型

text_region_mask (ndarray)

reorder_poly_edge(points)[源代码]

Get the respective points composing head edge, tail edge, top sideline and bottom sideline.

参数

points (ndarray) – The points composing a text polygon.

返回

The two points composing the head edge of text

polygon.

tail_edge (ndarray): The two points composing the tail edge of text

polygon.

top_sideline (ndarray): The points composing top curved sideline of

text polygon.

bot_sideline (ndarray): The points composing bottom curved sideline

of text polygon.

返回类型

head_edge (ndarray)

resample_line(line, n)[源代码]

Resample n points on a line.

参数
  • line (ndarray) – The points composing a line.

  • n (int) – The resampled points number.

返回

The points composing the resampled line.

返回类型

resampled_line (ndarray)

resample_sidelines(sideline1, sideline2, resample_step)[源代码]

Resample two sidelines to be of the same points number according to step size.

参数
  • sideline1 (ndarray) – The points composing a sideline of a text polygon.

  • sideline2 (ndarray) – The points composing another sideline of a text polygon.

  • resample_step (float) – The resampled step size.

返回

The resampled line 1. resampled_line2 (ndarray): The resampled line 2.

返回类型

resampled_line1 (ndarray)

class mmocr.datasets.pipelines.ToTensorNER[源代码]

Convert data with list type to tensor.

class mmocr.datasets.pipelines.ToTensorOCR[源代码]

Convert a PIL Image or numpy.ndarray to tensor.

mmocr.datasets.pipelines.sort_vertex(points_x, points_y)[源代码]

Sort box vertices in clockwise order from left-top first.

参数
  • points_x (list[float]) – x of four vertices.

  • points_y (list[float]) – y of four vertices.

返回

x of sorted four vertices. sorted_points_y (list[float]): y of sorted four vertices.

返回类型

sorted_points_x (list[float])

mmocr.datasets.pipelines.sort_vertex8(points)[源代码]

Sort vertex with 8 points [x1 y1 x2 y2 x3 y3 x4 y4]

utils

class mmocr.datasets.utils.HardDiskLoader(ann_file, parser, repeat=1)[源代码]

Load annotation file from hard disk to RAM.

参数

ann_file (str) – Annotation file path.

class mmocr.datasets.utils.LineJsonParser(keys=[])[源代码]

Parse json-string of one line in annotation file to dict format.

参数

keys (list[str]) – Keys in both json-string and result dict.

class mmocr.datasets.utils.LineStrParser(keys=['filename', 'text'], keys_idx=[0, 1], separator=' ', **kwargs)[源代码]

Parse string of one line in annotation file to dict format.

参数
  • keys (list[str]) – Keys in result dict.

  • keys_idx (list[int]) – Value index in sub-string list for each key above.

  • separator (str) – Separator to separate string to list of sub-string.

class mmocr.datasets.utils.LmdbLoader(ann_file, parser, repeat=1)[源代码]

Load annotation file with lmdb storage backend.

Read the Docs v: latest
Versions
latest
stable
Downloads
pdf
html
epub
On Read the Docs
Project Home
Builds

Free document hosting provided by Read the Docs.