0aa5d7be6d
[Model] Add SVTR framework and configs ( #1621 )
...
* [Model] Add SVTR framework and configs
* update
* update transform names
* update base config
* fix cfg
* update cfgs
* fix
* update cfg
* update decoder
* fix encoder
* fix encoder
* fix
* update cfg
* update name
2023-01-06 16:07:06 +08:00
b0557c2c55
[Transforms] SVTR transforms ( #1645 )
...
* rec transforms
* fix
* ut
* update docs
* fix
* new name
* fix
2023-01-06 16:04:20 +08:00
5dbacfe202
[Feature] Add svtr encoder ( #1483 )
...
* add svtr backbone
* update backbone
* fix
* apply comments, move backbone to encoder
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2023-01-04 16:03:21 +08:00
65e746eb3d
[UT] Add missing unit tests ( #1651 )
...
* update
* remove code
2022-12-30 12:01:14 +08:00
7e9f7756bc
[Feature] Add svtr decoder ( #1448 )
...
* add svtr decoder
* svtr decoder
* update
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2022-12-30 12:00:48 +08:00
9baf440d7a
[Feature] ConditionApply ( #1646 )
2022-12-28 11:53:32 +08:00
101f2b6eef
[Enhancement] Enhance FixInvalidPolygon, add RemoveIgnored transform ( #1632 )
...
* fix polygon_utils
* ut for poly_make_valid
* optimize crop_polygon
* FixInvalidPolygon, debug msg included
* add remove_pipeline_elements to utils
* enhance fixinvalidpolys
* fix transform_utils
* remove ignored
* RemoveIgnored
* add tests
* fix
* fix ut
* fix ut
2022-12-27 10:30:10 +08:00
419f98d8a4
[ASTER] Add ASTER decoder ( #1625 )
...
* add aster decoder
* aster decoder
* decoder
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2022-12-15 14:53:17 +08:00
0bd62d67c8
[ASTER] Add ASTER Encoder ( #1239 )
...
* [Docs] Limit extension versions (#1209 )
* loss
* fix
* [update] limit extension versions
* add aster encoder
* aster encoder
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2022-12-15 14:53:05 +08:00
ffe5237aa8
[Refactor] Refactor icdardataset metainfo to lowercase. ( #1620 )
2022-12-13 11:45:33 +08:00
3a0aa05d9c
[Feature] textspotting datasample ( #1593 )
...
* textspotting datasample
* rename
2022-12-06 14:03:32 +08:00
9ac9a227ec
[Improve] support head loss or postprocessor is None for only infer ( #1594 )
...
* support head loss or postprocessor is None for only infer
* default to None
2022-12-06 14:02:37 +08:00
5940d6bc9c
[Fix] textspotting ut ( #1599 )
2022-12-06 14:02:12 +08:00
a12c215e85
[Refactor] Refactor TPS ( #1240 )
...
* [Docs] Limit extension versions (#1209 )
* loss
* fix
* [update] limit extension versions
* refactor tps
* Update mmocr/models/textrecog/preprocessors/tps_preprocessor.py
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* Update mmocr/models/textrecog/preprocessors/tps_preprocessor.py
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* refine
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
2022-12-05 14:51:40 +08:00
b8c445b04f
[Fix] fix icdar data parse for text containing seperator ( #1587 )
...
* [Fix] fix icdar data parse for text containing seperator
* Update mmocr/datasets/preparers/parsers/base.py
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
2022-12-01 18:43:09 +08:00
cfce57ad87
[Feature] Add config generator ( #1552 )
...
* [Feature] Add config generator
* update icdar2013
* fix ut
* simplify design
* cfg generator
* update
* fix
2022-11-21 14:23:20 +08:00
37f3b88a05
[Feature] Add get_md5 ( #1553 )
...
* [Feature] Add get_md5
* fix
2022-11-18 15:22:47 +08:00
34e97abcb0
[Enhancement] Polish bbox2poly ( #1532 )
2022-11-15 19:40:12 +08:00
62ff782b71
[Fix] Fix ICDARTxtParser ( #1529 )
...
* fix
* fix comments
* fix
* fix
* fix
* delete useless
2022-11-15 19:18:14 +08:00
79a778689d
[Feature] Add SVT to dataset preparer ( #1521 )
...
* add svt to data preparer
* add svt parser test
2022-11-15 18:54:30 +08:00
f1dd437d8d
[Feature] support modified resnet structure used in oCLIP ( #1458 )
...
* support modified ResNet in CLIP and oCLIP
* update unit test for TestCLIPBottleneck; update docs
* Apply suggestions from code review
* fix
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
2022-11-03 17:54:15 +08:00
8864fa174b
[Feature] Add Dataset Preparer ( #1484 )
...
* add data preparer
* temporarily ignore data preparer test
* update
* fix comments
* update doc; add script to generate dataset zoo doc
* fix comments; update scripts
* apply comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* apply comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* coco parser
* fix comments
* add fileio tests
* fix test
* add tests for parsers and dumpers
* add test for data preparer
* fix a bug
* update icdar txt parser
* rename icdar txt parser
* fix comments
* fix test
* fix comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
Co-authored-by: liukuikun <641417025@qq.com>
2022-11-02 15:06:49 +08:00
dfc17207ba
[Vis] visualizer refine ( #1411 )
...
* visualizer refine
* updata docs
2022-10-09 12:45:17 +08:00
e801df3471
[CI] Fix CI ( #1365 )
2022-09-05 16:58:31 +08:00
db6ce0d95e
[Refactor] ocr.py ( #1344 )
...
* [Feature] Add BaseInferencer]
* [Feature] Add Det&Rec Inferencer
* [Feature] Add KIEInferencer
* [Feature] Add MMOCRInferencer
* [Refactor] update ocr.py
* update links
* update two links
* remove ocr.py
* move ocr.py and add loadfromndarray
Co-authored-by: xinyu <wangxinyu2017@gmail.com>
Co-authored-by: liukuikun <liukuikun@sensetime.com>
2022-08-31 22:56:24 +08:00
965f92f1e0
[Fix] Dataflow in mmdet wrapper ( #1335 )
...
* [Fix] Dataflow in mmdet wrapper
* fix typing
* fix ut
Co-authored-by: liukuikun <liukuikun@sensetime.com>
2022-08-31 19:03:59 +08:00
bf042f8267
[Feture] kie visualizer ( #1267 )
...
* kie visualizer
* add textspotting visualizer
* Fix
* fix
* Some fixes
Co-authored-by: gaotongxiao <gaotongxiao@gmail.com>
2022-08-29 20:13:06 +08:00
9b368fe45c
[Fix] Rename VeryDeepVGG to MiniVGG ( #1330 )
...
* [Fix] Rename VeryDeepVGG to MiniVGG
* update
2022-08-25 16:35:59 +08:00
b32412a9e9
[Refactor] MMEngine directory tree ( #1317 )
2022-08-25 11:46:04 +08:00
9bd5258513
[Refactor] Adapt to new dataflow ( #1305 )
...
* datasample->datasamples
* update rec data preprocessor
* rename datasamples
* update det preprocessor
* update metric
* update data_sample->data_samples in test
* update
* fix data preprocessor uts
* remove engine runner
* fix kie ut
* fix ut
* fix comments
* refactor evaluator ut
* apply comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* remove useless
* apply comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
* apply comments
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
Co-authored-by: Tong Gao <gaotongxiao@gmail.com>
2022-08-25 11:45:42 +08:00
1b5764b155
[Refactor] Rename base_xxx.py as base.py ( #1322 )
2022-08-25 11:20:42 +08:00
b81d58e70c
[Enhancement] Move dictionary ( #1320 )
2022-08-24 17:58:52 +08:00
c093c687a7
[Config] Refactor base config (part 3) ( #1316 )
...
* [Config] Refactor KIE
* fix ut
* fix edge classes
2022-08-24 17:30:05 +08:00
1860a3a3b6
[Feature] Use BCEWithLogitsLoss as many as possible for AMP ( #1309 )
...
* Use BCEWithLogitsLoss as many as possible for AMP
* fix
* Optimize DBNet
* fix docstr
* Use branch in dbhead, fix missing data_samples in textdethead
* fix
2022-08-23 19:17:27 +08:00
e760dcd1dd
[Enhancement] Support loading different lmdb datasets in LoadImageFromLMDB ( #1293 )
...
* [Enhancement] Support loading different lmdb datasets in LoadImageFromLMDB
* add docstr
2022-08-22 16:43:25 +08:00
d27b2fd84f
[Config] rec config refine ( #1306 )
...
* refine rec config
* fix mj-sub
2022-08-22 16:42:56 +08:00
240bf06ddd
[Fix] unit tests due to new config names ( #1303 )
2022-08-22 15:02:52 +08:00
c7a4298c32
[Fix] Replace mmcv.fileio with mmengine ( #1292 )
2022-08-19 16:53:14 +08:00
0d9b40706c
[Fix] Remove dependency on MMCV registry ( #1261 )
...
* [Fix] Remove dependency on MMCV registry
* fix
2022-08-19 11:19:07 +08:00
587566d2c2
[Fix] evaluator ut ( #1285 )
2022-08-16 17:34:07 +08:00
bcc245efd3
[Evaluator] MultiDatasetEvaluator ( #1250 )
...
* multi datasets evalutor
* fix comment
* fix typo
2022-08-16 16:48:04 +08:00
7b25b62c21
[Fix] Fix dictionary docstr and remove unncessary kwargs ( #1276 )
...
* [Fix] Fix dictionary docstr and remove unncessary kwargs
* fix
* fix
2022-08-11 11:14:17 +08:00
ef683206ed
[Api] vis hook and data flow api ( #1185 )
...
* vis hook and data flow api
* fix comment
* add TODO for merging and rewriting after MultiDatasetWrapper
2022-08-08 11:37:46 +08:00
80d85c129f
[Fix] Rename tests ( #1244 )
2022-08-01 15:28:27 +08:00
48cc575507
[Rename] Rename mmocr/data to mmocr/structures ( #1234 )
...
* rename mmocr/data to mmocr/structures
* update lint
2022-08-01 10:59:53 +08:00
717460055c
Revert "[update] Rename tests"
...
This reverts commit 0183a73b75
2022-07-27 17:52:35 +08:00
2cca103b93
[Fix] fix load error ( #1212 )
2022-07-26 16:53:11 +08:00
bc043101fe
[Fix] load error ( #1207 )
2022-07-26 10:48:10 +08:00
83e4fb10ee
[Fix] fix empty image and resize small image ( #1200 )
2022-07-25 19:21:09 +08:00
6f30020eec
[CI] Add CI ( #1176 )
...
* [CI] Add CI
* update init
* fix lint
* fix lint
* fix linting
* fix linting
* fix linting
* fix
* fix
* fix
* fix
* fix
* fix
* disable github ci
* fix
* Update .circleci/test.yml
Co-authored-by: Qing Jiang <mountchicken@outlook.com>
* fix
* fix
Co-authored-by: Qing Jiang <mountchicken@outlook.com>
2022-07-21 14:28:57 +08:00
Tong Gao
Yang Liu
Qing Jiang
liukuikun
Xinyu Wang
Wenqing Zhang