mirrors
/
mmocr
mirror of https://github.com/open-mmlab/mmocr.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
1,096 Commits
5 Branches
20 Tags
54 MiB
Commit Graph

322 Commits (7ef34c44076eed5df9b20312f9232e007f568ac8)

Author SHA1 Message Date
liukuikun b828d654a9 [Encoder] scanner encoder 2022-07-21 10:55:46 +08:00
gaotongxiao 460f068891 [DRRG] DRRG framework 2022-07-21 10:55:46 +08:00
gaotongxiao 9470821aa0 [DRRG] DRRG head 2022-07-21 10:55:46 +08:00
gaotongxiao e0992a7fae [DRRG] DRRG loss 2022-07-21 10:55:46 +08:00
gaotongxiao ed9e8d150c [DRRG] DRRG postprocessor 2022-07-21 10:55:46 +08:00
gaotongxiao 9db0941837 [SDMGR] Add SDMGR Head 2022-07-21 10:55:46 +08:00
gaotongxiao eaf7f6bf0c [SDMGR] postprocessor 2022-07-21 10:55:46 +08:00
gaotongxiao e23a2ef089 Add SDMGR Loss 2022-07-21 10:55:46 +08:00
liukuikun 622e65926e [Decoder]Robust scanner decoder 2022-07-21 10:55:45 +08:00
gaotongxiao 7490301877 Add KIE transforms 2022-07-21 10:55:45 +08:00
wangxinyu 52f0eefb2e [Transform] Add FixInvalidPolygon 2022-07-21 10:55:45 +08:00
gaotongxiao c5364f843d Add F-metric 2022-07-21 10:55:45 +08:00
xinyu 02a43d234e clean point utils todos 2022-07-21 10:55:45 +08:00
liukuikun b20bcc47b3 multi loop 2022-07-21 10:55:45 +08:00
xinyu 6ff567bb08 fix np warnings & fix nose_parameterized warnings 2022-07-21 10:55:45 +08:00
gaotongxiao 9f2fabc35a Add WildReceipt Dataset 2022-07-21 10:55:45 +08:00
gaotongxiao 09169f32ee Support skipping unknown tokens in dictionary 2022-07-21 10:55:45 +08:00
wangxinyu 4b185d3347 [Model] Add MobilenetV2 Backbone 2022-07-21 10:55:45 +08:00
wangxinyu ab6e897c6b [Utils] Migrate datasets/utils 2022-07-21 10:55:45 +08:00
wangxinyu ef98df8052 [Utils] Migrate core/evaluation/utils.py 2022-07-21 10:55:45 +08:00
jiangqing.vendor 83ec5726d6 Add RecogDatasets 2022-07-21 10:55:44 +08:00
liukuikun b955df9904 [Detector] refactor basedetector 2022-07-21 10:55:44 +08:00
jiangqing.vendor 0dc4fda545 Add IcdarDatset 2022-07-21 10:55:44 +08:00
liukuikun 988fea441b [Recognizer] refactor baserecognizer 2022-07-21 10:55:43 +08:00
gaotongxiao 0fb0d7cb1a Fix UT 2022-07-21 10:51:03 +08:00
liukuikun 41d9c741cd [Decoder] sar decoder 2022-07-21 10:51:03 +08:00
liukuikun 47771788f0 [Encoder] sar encoder 2022-07-21 10:51:03 +08:00
jiangqing.vendor 7be4dc1bca Refactor Parser 2022-07-21 10:51:03 +08:00
jiangqing.vendor 0bf1ce88c2 [FCENet] Add FCE Postprocessor 2022-07-21 10:51:03 +08:00
gaotongxiao cd4e520cb9 Add KIEDataSample 2022-07-21 10:51:03 +08:00
wangxinyu c0c0f4b565 [PSE] PSE Postprocessor 2022-07-21 10:51:03 +08:00
wangxinyu 4a04982806 [PSE] PSE Loss 2022-07-21 10:51:03 +08:00
gaotongxiao 0716c97cf6 Add BoundedScaleAspectJitter 2022-07-21 10:51:03 +08:00
wangxinyu 00ba46b5b9 [PSE] PSE Neck FPNF 2022-07-21 10:51:03 +08:00
wangxinyu 05990c58d9 [Refactor] PSE Head 2022-07-21 10:51:03 +08:00
gaotongxiao 490d6cd806 Fix MaskedCELoss 2022-07-21 10:51:03 +08:00
jiangqing.vendor fded755af2 [FCENet] Add FCENet loss 2022-07-21 10:51:03 +08:00
Mountchicken 17606c25fc add fce head 2022-07-21 10:51:02 +08:00
liukuikun 200899b2a0 [PAN Postprocessor]pan postprocessor 2022-07-21 10:51:02 +08:00
jiangqing.vendor 13920924ce [MASTER] Add master plugin 2022-07-21 10:51:02 +08:00
liukuikun b6e031666b [PAN] pan head 2022-07-21 10:51:02 +08:00
jiangqing.vendor 55c99dd0c1 [Update] Update TextDetRandomCropFlip 2022-07-21 10:51:02 +08:00
liukuikun 7c3789d64e [PANLoss] pan loss 2022-07-21 10:51:02 +08:00
liukuikun d636adeb1f fix ut 2022-07-21 10:51:02 +08:00
gaotongxiao 8f7c0e2977 Fix imgaug 2022-07-21 10:51:02 +08:00
gaotongxiao 21d0dd71dc [TextSnake] Refactor textsnake loss 2022-07-21 10:51:02 +08:00
wangxinyu bf7c738798 [TextSnake] TextSnake Neck 2022-07-21 10:51:02 +08:00
wangxinyu acd2bcc452 [TextSnake] TextSnake Postprocessor 2022-07-21 10:51:02 +08:00
wangxinyu f7731c43bd [TextSnake] TextSnake Head 2022-07-21 10:51:02 +08:00
liukuikun a353a28a1a fix test path 2022-07-21 10:51:02 +08:00
jiangqing.vendor a135580912 [MASTER] Add Master decoder 2022-07-21 10:51:01 +08:00
liukuikun f03ed3ce11 [Transform] RandomFlip 2022-07-21 10:51:01 +08:00
liukuikun d5a2d20574 [Enchance] ce ignore char 2022-07-21 10:51:01 +08:00
wangxinyu b44869059b [SATRN] SATRN Backbone 2022-07-21 10:51:01 +08:00
wangxinyu 401088913b [SATRN] SATRN Encoder 2022-07-21 10:51:01 +08:00
gaotongxiao 35e5138b5d Cleanup test tmp files in test crnn decoder 2022-07-21 10:51:01 +08:00
liukuikun dfe93dc7d2 [Transform] ScaleAspectJitter 2022-07-21 10:51:01 +08:00
gaotongxiao da175b44a4 [Fix] Fix RandomRotate 2022-07-21 10:51:01 +08:00
jiangqing.vendor 50f229d9fe [NRTR] NRTR Encoder 2022-07-21 10:51:01 +08:00
jiangqing.vendor 8614070e36 [NRTR] NRTR Decoder 2022-07-21 10:51:01 +08:00
jiangqing.vendor d41921f03d [NRTR] NRTR backbone 2022-07-21 10:51:01 +08:00
jiangqing.vendor 25e819f6bf [Fix] Fix TextDetRandomCrop 2022-07-21 10:51:01 +08:00
jiangqing.vendor b3b1ef146b [Fix] Check transform's unit test and visualize 2022-07-21 10:51:01 +08:00
jiangqing.vendor d859fcad1c [Update] Add toy dataset in 2.0 form for test 2022-07-21 10:51:00 +08:00
liukuikun 2a852f23b5 [Fix] fix hmean iou 2022-07-21 10:51:00 +08:00
gaotongxiao b585dbcdd7 Fix score field in DBPostprocessor 2022-07-21 10:51:00 +08:00
gaotongxiao a4952a6dd6 Fix RandomCrop 2022-07-21 10:51:00 +08:00
liukuikun 38eef984c2 [Refactor]CELoss 2022-07-21 10:51:00 +08:00
liukuikun f4a8e0f3a9 [Fix] fix metric and ut 2022-07-21 10:51:00 +08:00
gaotongxiao 8396b2014e Refactor FPNC 2022-07-21 10:51:00 +08:00
gaotongxiao 3a9f9e6b61 Handling a corner case in offset_polygon 2022-07-21 10:51:00 +08:00
gaotongxiao 32ef9cc3cf [DBNet] Add DBHead 2022-07-21 10:51:00 +08:00
gaotongxiao 7a66a84b64 [DBNet] Add DBPostProcessor 2022-07-21 10:51:00 +08:00
gaotongxiao cd3d173b18 [DBNet] Add DBLoss 2022-07-21 10:51:00 +08:00
gaotongxiao 747b2a14dc Refactor ResNet backbone 2022-07-21 10:51:00 +08:00
gaotongxiao 43c50eee82 fix loss 2022-07-21 10:51:00 +08:00
gaotongxiao 1e1da7b395 [DBNet] Add MaskedSmmothL1Loss, MasedBalancedBCELoss and MaskedDiceLoss 2022-07-21 10:50:59 +08:00
gaotongxiao 0f0f68baf1 Fix crop_polygon, Resize and poly2shapely 2022-07-21 10:50:59 +08:00
liukuikun bbbefaeb31 build LabelData in baseprocessor 2022-07-21 10:50:59 +08:00
liukuikun 7582fdea41 [Refactor] CTCLoss 2022-07-21 10:50:59 +08:00
liukuikun 3aae157aec [Refactor] crnn decoder 2022-07-21 10:50:59 +08:00
jiangqing.vendor f173cd3543 [Refactor] Refactor WordMetric and CharMetric 2022-07-21 10:50:59 +08:00
liukuikun 4fd048aa24 [Fix] fix base recog loss 2022-07-21 10:50:59 +08:00
liukuikun fe43259a05 [Refactor] train and test 2022-07-21 10:50:59 +08:00
wangxinyu ee48713a89 [Refactor] TextDetLocalVisualizer 2022-07-21 10:50:59 +08:00
jiangqing.vendor c78be99f6b [Refactor] Refactor TextRecogVisualizer 2022-07-21 10:50:59 +08:00
gaotongxiao 7e7a526f37 New Hmean-iou metric 2022-07-21 10:50:59 +08:00
liukuikun f47f3eff03 fix polygon type 2022-07-21 10:50:58 +08:00
liukuikun 05e31e09bc [Fix] base decoder forget passing dictionary 2022-07-21 10:50:58 +08:00
jiangqing.vendor 4706cc7eca [Refactor] Refactor SquareResizePad 2022-07-21 10:50:58 +08:00
liukuikun 2f4679e908 LoadAnnotations 2022-07-21 10:50:58 +08:00
gaotongxiao be30df5d50 Rename TextRecog directories 2022-07-21 10:50:58 +08:00
gaotongxiao f820a50752 Rename TextDet directories 2022-07-21 10:50:58 +08:00
liukuikun 84a61ba816 RescaleToHeight and PadToWidth 2022-07-21 10:50:58 +08:00
liukuikun a05e3f19c5 [Feature] TextRecogPostprocessor 2022-07-21 10:50:58 +08:00
liukuikun e8f57d6540 [Feature] PackDet/RecogInput 2022-07-21 10:50:58 +08:00
jiangqing.vendor 5dc791adbb [Refactor] Refactor randomscaling 2022-07-21 10:50:58 +08:00
jiangqing.vendor d2808e6b84 [refactor] Refactor Crop-Related Operations 2022-07-21 10:50:58 +08:00
gaotongxiao 7b6778c5d8 add torchvisionwrapper 2022-07-21 10:50:58 +08:00
liukuikun 0b5d2df310 [Refactor] BaseDecoder 2022-07-21 10:50:58 +08:00
liukuikun 6cd38a038f [Refactor] base recog loss 2022-07-21 10:50:57 +08:00
wangxinyu ac4eb34843 [Refactor] RandomCropFlip 2022-07-21 10:50:57 +08:00
gaotongxiao 79186b61ec ImgAug: ignores->ignored 2022-07-21 10:50:57 +08:00
wangxinyu 178030bad6 [Refactor] Refactor transform.RandomRotate 2022-07-21 10:50:57 +08:00
jiangqing.vendor f29853d9cd [Feature] Add Resize 2022-07-21 10:50:57 +08:00
gaotongxiao 6478499073 add mode to rescale_polygon(s) 2022-07-21 10:50:57 +08:00
gaotongxiao df2f7b69db Add recognition data migrator 2022-07-21 10:50:57 +08:00
gaotongxiao 7b09da485b [Fix] Support textrecog, add more tests 2022-07-21 10:50:57 +08:00
liukuikun 0f041d4250 [Fix] updata LabelData import 2022-07-21 10:50:57 +08:00
gaotongxiao f6b72b244b Add ImgAug and tests 2022-07-21 10:50:57 +08:00
gaotongxiao cb85f857aa Add BaseTextDetPostProcessor 2022-07-21 10:50:57 +08:00
gaotongxiao f0c6d44ce8 Add dump_ocr_data 2022-07-21 10:50:56 +08:00
liukuikun 98bc90bd1c [Feature] TextRecogSample 2022-07-21 10:50:56 +08:00
liukuikun c920edfb3a [Refactor] split labelconverter to Dictionary 2022-07-21 10:50:56 +08:00
liukuikun c47c5711c1 [Feature] TextDetSample 2022-07-21 10:50:56 +08:00
wangxinyu.vendor 41c1671e7b Refactor PyramidRescale 2022-07-21 10:50:56 +08:00
gaotongxiao b5fc589320 Add CI 2022-07-21 10:50:55 +08:00
gaotongxiao 69e6c80558 move tests to old_tests, add empty test folders 2022-07-21 10:50:55 +08:00
Tong Gao 1ad4caa7cb
[Enhancement] Allow choosing metrics to report in text recognition tasks (#989) 
* [Enhancement] Allow choosing metrics to report in text recognition tasks

* add acc support

* refine docstr
 2022-05-05 16:10:38 +08:00
Jianquan Ye 061d353b2a
[Model] Add MASTER (#807) 
* fix #794: add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* Fix linting

* after git rebase main

* after git rebase main

* fix conflict add MASTER

* fix conflict add MASTER

* after git rebase main

* fix conflict add MASTER

* fix conflict add MASTER

* fix conflict add MASTER

* after git rebase main

* add GCAModule to plugins

* coexist master and master_old

* fix merge mmocr 0.5.0 conflict

* fix lint error

* update

* [fix] remove remains in __init__

* [update] update code in review

* update readme for master

* Add docstr to MasterDecoder, refined MasterDecoder, remove MASTERLoss

* Unify the output length of MasterDecoder in train and test mode; add test for it, remove MasterLoss

* update readme

* update

* update metafile,README,demo/README,config,ocr.py

* Update mmocr/utils/ocr.py

* update

Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
Co-authored-by: Mountchicken <mountchicken@outlook.com>
 2022-05-05 16:06:15 +08:00
Xinyu Wang fbc138d2d3
[Feature] Add DBNet++ (#973) 
* add dbnet++

* fix docstring

* fix some param names

* fix

* fix docstring

* add init

* add doc; remove configs

* add dbnet++ to readme

* fix readme

* update config

* update readme

* update readme

* update ocr.py

* update metafile.yml

* update readme

* update readme

* move to dbnetpp

* fix paths

* fix head level

* fix typo

* update demo.md

* Update configs/textdet/dbnetpp/README.md

Co-authored-by: Tong Gao <gaotongxiao@gmail.com>

* fix typo

* fix link
 2022-05-05 14:39:20 +08:00
Tong Gao 6b180db93d
[Enhancement] Automatically report mean scores when applicable (#995) 2022-05-05 12:57:19 +08:00
Hongbin Sun a2d741b8a7
[Feature] Add labelme converter for textdet and textrecog (#972) 
* add labelme converter

* move to common

* add labelme sample annos

* add doc

* remove useless field generated by labelme to reduce size

* add recog_format option; add skip ignored instances while cropping

* set warp as false by default

* update doc

* fix typo

Co-authored-by: xinke-wang <wangxinyu2017@gmail.com>
Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>
 2022-05-03 17:28:22 +08:00
Qing Jiang c91763ad09
[Fix] update label.lmdb (#991) 
* loss

* fix

* hotfix
 2022-04-29 23:26:32 +08:00
Qing Jiang 92ef554a82
[Feature] Add recog2lmdb and new toy dataset files (#979) 
* loss

* fix

* add img2lmdb and test files

* update

* add reference

* fix lint

* fix typo

* use total_numer instead to fit mmocr's lmdbloader

* reorganize and update

* fix lint

* update test file

* refactor and update

* fix test

* update doc in tools

* fix lint

* update old lmdb test file

* update

* mask the unittest for recog2lmdb and use json format for label_only

* remove if __name__

* fix case, doc, typo, formats

* fix typos

* fix docs and variable names

* Apply suggestions from code review

Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>

* update test_loader.py and fix a bug

Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>
 2022-04-29 22:30:36 +08:00
Qing Jiang d8549ade85
[Feature] Support loading images and labels from lmdb file (#982) 
* loss

* fix

* add LoadImageFromLMDB

* update

* new version

* add json

* fix bug

* Compatible with older versions

* modify test_loader

* add docstring and update test_loader

* fix a bug

* fix docstring and update doc

* update

* Update mmocr/datasets/utils/backend.py

Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>

* Update docs/en/tutorials/dataset_types.md

Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>

* fix lint

Co-authored-by: Xinyu Wang <45810070+xinke-wang@users.noreply.github.com>
 2022-04-29 15:46:36 +08:00
Tong Gao 064a2b8c45
[Feature] Support computing mean scores in UniformConcatDataset (#981) 
* Get avg results in UniformConcatDataset

* add docstr

* Fix

* fix test

* fix typo
 2022-04-29 14:48:21 +08:00
Tong Gao 0c8fa52b22
[Fix] Fix ImgAug transform (#949) 
* [Fix] Fix bbox case in ImgAug

* Finialize the augmentation and add tests
 2022-04-28 22:52:11 +08:00
Tong Gao 888f700134
[Enhancement] Support dynamic threshold range in eval_hmean (#962) 
* [Enhancement] Support dynamic threshold range in eval_hmean

* upgrade textdetdataset, add deprecate warning
 2022-04-22 17:07:36 +08:00
garvan2021 b11fa80d82
[Fix] correct meta key (#926) 
* correct meta key

* fix test metakey
 2022-04-18 09:09:09 +08:00
garvan2021 c79a62487d
[Feature] Add Tesserocr Inference (#814) 
* append tesserocr to requirements list, but may encounter build error at windows platform

* simply save

* 2022.3.4

* opencv-python==4.5.5 can cause cv2.error when print_result=True

* append MMOCR.tesseract_det_inference()

* argument check append

* fix lint error

* update commentary

* lint fix

* requirement remove opencv

* handle tessdata problem

* support tesseract recognition

* fix some bugs

* fix imshow bug

* support batch mode(fake)

* modify annotation

* refactor BaseRecognizer for show_result

* append pytest

* Mock tesseract

* Fix test

* remove \n from Tesseract

* normalize text score

* update docs
 2022-03-28 22:50:53 +08:00
Tong Gao 37833ada4b
[Fix] Support loading space character from dict file (#854) 
* [Feature] Support loading dict file with space character

* fix tests

* clean up \\r in tests

* add DICT37 and DICT91

* update docstr
 2022-03-22 17:44:32 +08:00
JiangQing 72f9454573
[Refactor] Refactor ResNets (#809) 2022-03-18 10:36:03 +08:00
JiangQing 680dff373e
[Feature] Support jsonl in recognition converter (#844) 2022-03-18 09:22:32 +08:00
Hongbin Sun 402e8f1162
[Feature] Support loading annotation file with petrel/http backend (#793) 
* add petrel backend for ann_file loading

* support icdar dataset

* add http backend

* name back

* fix pytest

* support load lmdb file on ceph

* refactor

* add close for lmdb

* add log and warning msg

* rename

* typo

* update

* fix
 2022-03-07 19:18:02 +08:00
Tong Gao fb77352eb2
[Fix] Standardize the type of torch.device in ocr.py (#800) 2022-03-03 14:18:33 +08:00
Tong Gao 3110ab7863
[Enhancement] Add windows CI (#790) 
* [Enhancement] Add windows CI

* [Enhancement] Add windows CI

* update

* update

* update

* [Fix] using assert will keep lmdb file opend and fail to cleanup in test_loader.py

* [Fix] map size should be small on windows in lmdb_util.py

* [Fix] Fix some bugs

* [Fix] Fix some bugs

* [Fix] Fix some bugs

* remove comments & fix bugs

Co-authored-by: Mountchicken <mountchicken@outlook.com>
 2022-03-02 10:34:15 +08:00
Tong Gao 5582e17feb
[Enhancement] Automatically infer device for inference if not speicifed (#781) 
* [Enhancement] Automatically infer device for inference if not speicifed

* update docs

* fix unit test
 2022-02-16 20:42:47 +08:00
Tong Gao 0f5c7d38f8
[Fix] Disable worker in the dataloader of gpu unit test 2022-02-16 20:42:22 +08:00
Tong Gao c466e6f82d
[Enhancement] Upgrade isort (#771) 
* [Enhancement] Upgrade isort

* Fix files
 2022-02-09 23:22:14 +08:00
liukuikun d871eed79e
[Enhancement] Add setup multi-processing both in train and test. (#757) 2022-01-26 20:34:19 +08:00
yuexy 82488024db [Feature] Add bezier_to_polygon and sort_points to box_util (#703) 2022-01-17 15:26:55 +08:00
Hongbin Sun fb1892a1ae
[Fix] fix bug of show_results in single_gpu_test (#667) 
* move single_gpu_test to mmocr/apis

* fix show result in single_gpu_test

* add unittest

* add docstring

* add unittest for kie

* avoid re-define

* remove skipif

* add skip

* fix conflict

* fix with comment

* add note in readme

* update
 2021-12-23 03:30:01 +00:00
Hongbin Sun f8dfbd4177
[Fix] Support both ConcatDataset and UniformConcatDataset (#675) 
* support UniformConcatDataset

* update

* rm useless

* handle 2d-list datasets
 2021-12-22 12:32:02 +00:00
Tong Gao 9104667112
[Model] Full ABINet Framework (#651) 
Co-authored-by: liukuikun <24622904+Harold-lkk@users.noreply.github.com>
 2021-12-15 11:21:54 +08:00
Hongbin Sun a4237ad568
[Enhancement] Support muitiple pipelines for different datasets (#657) 
* support multi pipelines

* support different pipelines

* fix

* fix replace_image_to_tensor

* remove redundant check

* update

* add copyright

* not in-place modify
 2021-12-14 12:29:43 +00:00
Hongbin Sun 7874531db3
[Refactor] refactor textdet postprocess (#640) 
* refactor textdet postprocess using build_postprocessor

* update docstring

* update docstring

* update postprocess docstring

* compatible with previous configs

* rename to BaseTextDetector

* rename to DRRGPostprocessor

* fix

* update warning

* fix

* fix conflict

* rename back

* rename warpper to utils

* fix doc
 2021-12-14 17:35:44 +08:00
Jianyong Chen 5caa945a8d
fix #614: textsnake targets (#660) 
* fix #614: textsnake targets

* fix lint

* add textsnake_targets test cases

* init with eps

* fix test coverage
 2021-12-13 17:45:30 +08:00
Hongbin Sun c6c4acd39b
[Fix] fix type check (#650) 
* fix check type

* add notice
 2021-12-09 12:45:59 +08:00
Hongbin Sun fec986aba0
[Enhancement] Use shape from loaded image (#652) 
* use shape from loaded image

* complete unittest

* add params as gate

* add comment

* add assert
 2021-12-09 01:34:09 +00:00