From 941fd5f465baf555578aaa5364ee11d175a73739 Mon Sep 17 00:00:00 2001 From: Topdu <784990967@qq.com> Date: Tue, 10 May 2022 08:18:31 +0000 Subject: [PATCH] [doc] add svtr algorithm_overview --- doc/doc_ch/algorithm_overview.md | 6 +++--- doc/doc_en/algorithm_overview_en.md | 2 ++ 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/doc/doc_ch/algorithm_overview.md b/doc/doc_ch/algorithm_overview.md index 313ef9b1..6227a214 100755 --- a/doc/doc_ch/algorithm_overview.md +++ b/doc/doc_ch/algorithm_overview.md @@ -44,7 +44,7 @@ 在CTW1500文本检测公开数据集上，算法效果如下： |模型|骨干网络|precision|recall|Hmean|下载链接| -| --- | --- | --- | --- | --- | --- | +| --- | --- | --- | --- | --- | --- | |FCE|ResNet50_dcn|88.39%|82.18%|85.27%|[训练模型](https://paddleocr.bj.bcebos.com/contribution/det_r50_dcn_fce_ctw_v2.0_train.tar)| **说明：** SAST模型训练额外加入了icdar2013、icdar2017、COCO-Text、ArT等公开数据集进行调优。PaddleOCR用到的经过整理格式的英文公开数据集下载： @@ -65,6 +65,7 @@ - [x] [NRTR](./algorithm_rec_nrtr.md) - [x] [SAR](./algorithm_rec_sar.md) - [x] [SEED](./algorithm_rec_seed.md) +- [x] [SVTR](./algorithm_rec_svtr.md) 参考[DTRB](https://arxiv.org/abs/1904.01906)[3]文字识别训练和评估流程，使用MJSynth和SynthText两个文字识别数据集训练，在IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE数据集上进行评估，算法效果如下： @@ -82,6 +83,7 @@ |NRTR|NRTR_MTB| 84.21% | rec_mtb_nrtr | [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar) | |SAR|Resnet31| 87.20% | rec_r31_sar | [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/rec/rec_r31_sar_train.tar) | |SEED|Aster_Resnet| 85.35% | rec_resnet_stn_bilstm_att | [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.1/rec/rec_resnet_stn_bilstm_att.tar) | +|SVTR|SVTR-Tiny| 89.25% | rec_svtr_tiny_none_ctc_en | [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/rec_svtr_tiny_none_ctc_en_train.tar) | @@ -90,5 +92,3 @@ 已支持的端到端OCR算法列表（戳链接获取使用教程）： - [x] [PGNet](./algorithm_e2e_pgnet.md) - - diff --git a/doc/doc_en/algorithm_overview_en.md b/doc/doc_en/algorithm_overview_en.md index 0cee8f4a..5c5c0306 100755 --- a/doc/doc_en/algorithm_overview_en.md +++ b/doc/doc_en/algorithm_overview_en.md @@ -58,6 +58,7 @@ Supported text recognition algorithms (Click the link to get the tutorial): - [x] [NRTR](./algorithm_rec_nrtr_en.md) - [x] [SAR](./algorithm_rec_sar_en.md) - [x] [SEED](./algorithm_rec_seed_en.md) +- [x] [SVTR](./algorithm_rec_svtr.md) Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation result of these above text recognition (using MJSynth and SynthText for training, evaluate on IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE) is as follow: @@ -75,6 +76,7 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r |NRTR|NRTR_MTB| 84.21% | rec_mtb_nrtr | [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar) | |SAR|Resnet31| 87.20% | rec_r31_sar | [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/rec/rec_r31_sar_train.tar) | |SEED|Aster_Resnet| 85.35% | rec_resnet_stn_bilstm_att | [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.1/rec/rec_resnet_stn_bilstm_att.tar) | +|SVTR|SVTR-Tiny| 89.25% | rec_svtr_tiny_none_ctc_en | [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/rec_svtr_tiny_none_ctc_en_train.tar) | -- GitLab