method: Unis_OCR2021-05-08

Authors: Jie.Li(李杰),BaoLin.Zhang(张保林),KeJie.Liu(刘克捷),Yuan.Hu(胡源)

Affiliation: UNISINSIGHT

Description: First, the training method is based on crnn framework, which takes SE-ResNet with multi-scale features as the backbone, and uses the BiLSTM and attention mechanism to integrate multiple model results. At the same time, the training data used include LSVT, ReCTS, RCTW, ArT and other public free data.

method: DH_OCR2021-05-08

Authors: Qiang Zeng(曾强),Zhaolin You(游照林),Yuanyuan Chen(陈媛媛),Jianping Xiong(熊剑平)

Affiliation: ZHEJIANG DAHUA TECHNOLOGY CO.,LTD

Description: Our training data included ReCTS, LSVT, RCTW, ART and some high-quality artificial synthetic data. We used the CRNN framework for text recognition, and different structures of multi-scale feature extraction backbone such as SA-ResNet, SE-ResNet were used.We used an efficient shuffle attention method which combine spatial attention with channel attention.Meanwhile we used multi-model fusion to predict the final result.

method: Sogou_OCR2020-12-23

Authors: Jianzhong Xu, Hailong Wang,Long Ma

Description: Our method is based on crnn framework. We use SE-ResNet with multi-scale feature as the backbone, the extracted feature is fused based on a two-layer transformer unit. Meanwhile, we introduce squeeze-and-excitation and relative position encodings to transformer. Our training datasets consist 20 million samples, including ReCTS, Art. Model with the same architecture have been deployed online for Sogou Input Text Scanning.

Ranking Table

Description Paper Source Code
DateMethodResult
2021-05-08Unis_OCR97.01%
2021-05-08DH_OCR97.00%
2020-12-23Sogou_OCR96.84%
2020-10-28Eleme-AI-V196.81%
2020-10-20PingAn_VisualComputing96.62%
2019-10-15MCEM_v3-iFLYTEK95.75%
2019-04-30SANHL_v195.55%
2020-10-12transformer_v195.10%
2019-10-30Encoder_Decoder_v195.09%
2019-04-30Tencent-DPPR Team94.86%
2019-04-30TPS-ResNet v194.77%
2019-04-30MCEM v294.22%
2019-04-30VOCR92.50%
2019-04-29CLTDR92.33%
2019-04-30Task2-re591.45%
2019-04-30HUST_Reg89.01%
2019-04-29ReCTS_HWY87.65%
2019-04-29Ssubm190429_nonchar_thres0082.17%
2019-04-30LCT_OCR (中国科学院信息工程研究所)80.96%
2019-04-27resnet101lstm78.26%
2019-04-30Scene text detection of polar coordinate regression73.79%
2019-04-30ECUST_dpc66.46%
2019-04-30jxl_ocr66.46%
2019-04-30ECUST_OCR63.96%

Ranking Graphic