4b887676a3
[Fix] MJSynth & SynthText Dataset Preparer config ( #1805 )
...
* [Fix] MJSynth
* update
* fix
* fix
2023-03-28 16:20:24 +08:00
bb591d2b1b
[Config] CTW1500 ( #1814 )
2023-03-28 10:40:53 +08:00
59d89e10c7
[Doc] Dataset ( #1782 )
...
* [Doc] Dataset
* fix
* update
* update
2023-03-27 12:47:01 +08:00
73df26d749
[Enhancement] Accepts local-rank in train.py and test.py ( #1806 )
...
* [Enhancement] Accepts local-rank
* add
* update
2023-03-27 10:34:54 +08:00
f47cff5199
[Bug] if dst not exists, when move a single file may raise a file not exist error. ( #1803 )
2023-03-24 14:14:45 +08:00
c886936117
[Enhancement] decouple batch_size to det_batch_size, rec_batch_size and kie_batch_size in MMOCRInferencer ( #1801 )
...
* decouple batch_size to det_batch_size, rec_batch_size, kie_batch_size and chunk_size in MMOCRInferencer
* remove chunk_size parameter
* add Optional keyword in function definitions and doc strings
* add det_batch_size, rec_batch_size, kie_batch_size in user_guides
* minor formatting
2023-03-24 11:06:03 +08:00
22f40b79ed
[Enhancement] Add MMOCR tutorial notebook ( #1771 )
...
* [Enhancement] Add MMOCR tutorial notebook
* revert
* [Enhancement] Add MMOCR tutorial notebook
* clear output
* update
2023-03-22 16:46:16 +08:00
1a379f2f1b
[Fix] Clear up some unused scripts ( #1798 )
2023-03-22 14:00:55 +08:00
d0dc90253a
[Dataset Preparer] MJSynth ( #1791 )
...
* finialize
* finialize
2023-03-22 10:10:46 +08:00
6d9582b6c7
[Docs] Fix quick run ( #1775 )
...
* Fix quick run
* fix ut
* debug
* fix
* fix ut
2023-03-22 10:09:54 +08:00
e0707bf5f2
[Fix] Synthtext metafile ( #1788 )
...
* [Fix] Synthtext metafile
* spotting
* fix
2023-03-21 11:30:21 +08:00
ae252626d3
[Docs] Fix some deadlinks in the docs ( #1469 )
...
* [Docs] Fix some deadlinks in the docs
* intersphines
2023-03-20 11:35:04 +08:00
d80df99037
[Fix] Add pse weight to metafile ( #1787 )
2023-03-20 11:12:13 +08:00
506f7d296e
[Fix] Test svtr_small instead of svtr_tiny ( #1786 )
2023-03-20 11:09:53 +08:00
9caacc76ee
[Enhancement] Deprecate file_client_args and use backend_args instead ( #1765 )
...
* tmp commit
* remove
2023-03-20 10:33:20 +08:00
63a6ed4e6c
[Fix] Place dicts to .mim ( #1781 )
...
* [Fix] Place dicts to .mim
* fix
* mmengine
2023-03-20 10:32:59 +08:00
c6580a48c1
[Dataset Preparer] SynthText ( #1779 )
...
* [Dataset] Support Synthtext
* update
* update
* finalize setting
* fix
* textrec
* update
* add fake magnet obtainer
* update rec
* update
* sample_ann
2023-03-17 14:34:12 +08:00
7ef34c4407
[Doc] add opendatalab download link ( #1753 )
...
* add opendatalab link
* fix
* fix
* ip
---------
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2023-03-14 15:55:53 +08:00
47f54304f5
[Enhancement] Make lanms-neo optional ( #1772 )
...
* [Enhancement] Make lanms-neo optional
* fix
* rm
2023-03-10 15:33:12 +08:00
465316f193
[Docs] Mark projects in docs ( #1766 )
...
* [Docs] Mark projects in docs
* fix
* fix
* fix
* fix
2023-03-10 15:09:37 +08:00
590af4b5e8
[Doc] Remove LoadImageFromLMDB from docs ( #1767 )
2023-03-10 14:52:16 +08:00
a58c77df80
[Docs] FAQ ( #1773 )
2023-03-10 14:51:57 +08:00
e9b23c56ad
Cherry Pick #1205 ( #1774 )
...
* [Bug fix] box points ordering (#1205 )
* fix sort_points
* remove functools
* add test cases
* add descriptions for coordinates
* del
Co-authored-by: xinyu <wangxinyu2017@gmail.com>
* fix
---------
Co-authored-by: liferecords <yjmm10@yeah.net>
Co-authored-by: xinyu <wangxinyu2017@gmail.com>
2023-03-10 14:51:40 +08:00
75c06d34bb
[Dataset Preparer] Add SCUT-CTW1500 ( #1677 )
...
* update metafile and download
* update parser
* updata ctw1500 to new dataprepare design
* add lexicon into ctw1500 textspotting
* fix
---------
Co-authored-by: liukuikun <641417025@qq.com>
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2023-03-08 17:32:00 +08:00
bfb36d81b3
fix ( #1764 )
2023-03-07 20:25:48 +08:00
45a8d89fb9
Bump version to 1.0.0rc6 ( #1763 )
...
* Bump version to 1.0.0rc6
* fix
* update changelog
* fix
* fix
2023-03-07 20:22:54 +08:00
d56155c82d
[Feature] Support lmdb format in Dataset Preparer ( #1762 )
...
* [Dataset Preparer] Support lmdb format
* fix
* fix
* fix
* fix
* fix
* readme
* readme
2023-03-07 20:08:25 +08:00
33cbc9b92f
[Docs] Inferencer docs ( #1744 )
...
* [Enhancement] Support batch visualization & dumping in Inferencer
* fix empty det output
* Update mmocr/apis/inferencers/base_mmocr_inferencer.py
Co-authored-by: liukuikun <24622904+Harold-lkk@users.noreply.github.com>
* [Docs] Inferencer docs
* fix
* Support weight_list
* add req
* improve md
* inferencers.md
* update
* add tab
* refine
* polish
* add cn docs
* js
* js
* js
* fix ch docs
* translate
* translate
* finish
* fix
* fix
* fix
* update
* standard inferencer
* update docs
* update docs
* update docs
* update docs
* update docs
* update docs
* en
* update
* update
* update
* update
* fix
* apply sugg
---------
Co-authored-by: liukuikun <24622904+Harold-lkk@users.noreply.github.com>
2023-03-07 18:52:41 +08:00
cc78866ed7
[Fix] SPTS readme ( #1761 )
2023-03-07 18:41:37 +08:00
f250ea2379
[Fix] Fix wrong ic13 textspotting split data; add lexicons to ic13, ic15 and totaltext ( #1758 )
...
* [Fix] Fix wrong ic13 textspotting split data; add lexicons to ic13, ic15 and totaltext
* [Fix] Fix wrong ic13 textspotting split data; add lexicons to ic13, ic15 and totaltext
* update
2023-03-07 14:23:00 +08:00
5685bb0f38
[Enhancement] configs for regression benchmark ( #1755 )
2023-03-07 14:19:22 +08:00
5670695338
[SPTS] train ( #1756 )
...
* [Feature] Add RepeatAugSampler
* initial commit
* spts inference done
* merge repeat_aug (bug in multi-node?)
* fix inference
* train done
* rm readme
* Revert "merge repeat_aug (bug in multi-node?)"
This reverts commit 393506a97c2089b02b48
2023-03-07 14:18:01 +08:00
81fd74c266
[Enchance] Put all registry into registry.py ( #1760 )
2023-03-07 14:17:06 +08:00
47f7fc06ed
[Feature] Support batch augmentation through BatchAugSampler ( #1757 )
...
* [Fix] RepeatAugSampler -> BatchAugSampler
* update docs
2023-03-07 11:29:53 +08:00
82f81ff67c
[Refactor] Refactor data converter and gather ( #1707 )
...
* Refactor dataprepare, abstract gather, packer
* update ic13 ic15 naf iiit5k cute80 funsd
* update dataset zoo config
* add ut
* finsh docstring
* fix coco
* fix comment
2023-03-03 15:27:19 +08:00
3aa9572a64
Remove outdated resources in demo/ ( #1747 )
...
* Remove outdated resources in demo/
* update
2023-02-28 22:09:02 +08:00
62d440fe8e
[Feature] add a new argument font_properties to set a specific font file in order to draw Chinese characters properly ( #1709 )
...
* [Feature] add new argument font_properties to set specific font file in order to draw Chinese characters properly
* update the minimum mmengine version
* add docstr
2023-02-27 14:25:49 +08:00
0894178343
Fixing Bounding Box format for Text Detection isn't specified ( #1734 )
...
* Update dataset.md
* fix
---------
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2023-02-22 10:51:54 +08:00
7cfd412ce7
[Docs] Fix CharMetric P/R wrong definition ( #1740 )
2023-02-22 10:29:10 +08:00
280a89c18e
[Fix] bezier_to_polygon -> bezier2polygon ( #1739 )
2023-02-21 19:16:17 +08:00
6eaa0673f7
[Fix] COCOTextv2 config ( #1737 )
2023-02-20 18:43:20 +08:00
9b0f1da1e7
[Fix] icdar textrecog ann parser skip ignore data ( #1708 )
2023-02-17 16:21:38 +08:00
37c5d371c7
Fix some browse dataset script bugs and draw textdet gt instance with ignore flags ( #1701 )
...
* [Enhancement] textdet draw gt instance with ignore flags
* [Fix] 明确key值定义，防止后续使用img_path时得到lmdb格式中的img_key导致图片无法读取
* [Fix] fix five browse_dataset.py script bugs
* [Fix] fix some pr problems
* [Fix] keep img_path attribute
* [Fix] 防止width很大text很小时，font_size过大显示不全（做keep_ratio的resize然后padding到固定尺寸时可能出现此类情况）
2023-02-17 15:40:24 +08:00
e9bf689f74
[Enhancement] Support batch visualization & dumping in Inferencer ( #1722 )
...
* [Enhancement] Support batch visualization & dumping in Inferencer
* fix empty det output
* Update mmocr/apis/inferencers/base_mmocr_inferencer.py
Co-authored-by: liukuikun <24622904+Harold-lkk@users.noreply.github.com>
---------
Co-authored-by: liukuikun <24622904+Harold-lkk@users.noreply.github.com>
2023-02-17 12:40:09 +08:00
1127240108
[Feature] Support auto import modules from registry. ( #1731 )
...
* [Feature] Support auto import modules from registry.
* limit mmdet version
* location parrent dir if it not exist
2023-02-17 10:28:34 +08:00
df0be646ea
[Enhancement] Speedup formatting by replacing np.transpose with torch.permute ( #1719 )
2023-02-16 14:14:03 +08:00
f820470415
[Feature] Rec TTA ( #1401 )
...
* Support TTA for recognition
* updata readme
* updata abinet readme
* updata train_test doc for tta
2023-02-16 10:27:07 +08:00
7cea6a6419
[Enchancement] Only keep meta and state_dict when publish model ( #1729 )
...
* Only keep meta and state_dict when publish model
* simpy code
2023-02-15 19:45:12 +08:00
3240bace4a
docs: fix the head show in readme ( #1727 )
2023-02-15 16:07:24 +08:00
b21d2b964a
docs: Add twitter discord medium youtube link ( #1724 )
2023-02-15 11:10:41 +08:00
Tong Gao
Kevin Wang
Hugo Tong
jorie-peng
Qing Jiang
liukuikun
EuanHoll
double22a
vansin