- Task 1 - Text Localization
- Task 2 - Script identification
- Task 3 - Joint text detection and script identification
- Task 4 - End-to-End text detection and recognition
method: Tencent-DPPR Team2019-06-04
Authors: Sicong Liu, Haoxi Li, Haibo Qin, Ben Xu, Chunchao Guo, Longhuang Wu, Shangxuan Tian, Hongfa Wang, Hongkai Chen, Qinglin lu, Chun Yang, Xucheng Yin, Lei Xiao
Description: We are from Tencent-DPPR (Data Platform Precision Recommendation) Team. We first recognize text lines and their character-level language types using ensemble results of several recognition models, which based on CTC/Seq2Seq and CNN with self-attention/RNN. After that, we identify the language types of recognized results based on statics of MLT-2019 and Wikipedia corpus.
method: Tencent-DPPR Team (Method_v0.3)2019-06-04
Authors: Sicong Liu, Haoxi Li, Haibo Qin, Ben Xu, Chunchao Guo, Longhuang Wu, Shangxuan Tian, Hongfa Wang, Hongkai Chen, Qinglin lu, Chun Yang, Xucheng Yin, Lei Xiao
Description: We are from Tencent-DPPR (Data Platform Precision Recommendation) Team. We first recognize text lines using ensemble results of several recognition models, which based on CTC/Seq2Seq and CNN with self-attention/RNN. After that, we identify the language types of recognized results based on statics of MLT-2019 and Wikipedia corpus.
method: Multi_scale_v12019-09-29
Authors: Wuheng Xu, Changxu Cheng, Bohan Li
Description: We used area block feature information using images on multiple scales.This model has 4 scales and 8 branches.We also used some data augments and improved ROI pooling.Finally, we used three training sets(mlt17, mlt19_train, mlt19_val).
| Date | Method | Script classification accuracy | |||
|---|---|---|---|---|---|
| 2019-06-04 | Tencent-DPPR Team | 94.03% | |||
| 2019-06-04 | Tencent-DPPR Team (Method_v0.3) | 93.16% | |||
| 2019-09-29 | Multi_scale_v1 | 91.71% | |||
| 2019-06-03 | CNN-Based Classifier | 91.66% | |||
| 2019-05-28 | GSPA_HUST | 91.02% | |||
| 2019-06-03 | SCUT-DLVC-Lab | 90.97% | |||
| 2019-06-04 | TPS-ResNet | 90.90% | |||
| 2019-06-04 | conv-transformer | 90.88% | |||
| 2019-06-04 | TH-DL-v2 | 90.70% | |||
| 2019-05-30 | conv-transformer | 90.30% | |||
| 2019-06-03 | TH-DL-v1 | 90.27% | |||
| 2019-05-27 | Tencent-DPPR Team (Method_v0.2) | 90.07% | |||
| 2019-05-27 | Tencent-DPPR Team (Method_v0.1) | 89.75% | |||
| 2019-05-28 | GS_HUST | 89.07% | |||
| 2019-05-28 | TH-ML | 88.85% | |||
| 2019-05-28 | MultiScale_HUST | 88.64% | |||
| 2019-05-27 | baseline2 | 88.54% | |||
| 2019-06-02 | Conv_Attention | 88.41% | |||
| 2019-05-27 | TH-DL | 88.09% | |||
| 2019-05-30 | cold | 87.98% | |||
| 2019-05-27 | NXB OCR | 84.88% | |||
| 2019-06-03 | NXB OCR | 84.86% | |||
| 2019-06-03 | Res_MUL_SPP_BUPT | 71.31% | |||
| 2019-06-02 | Res_SPP_BUPT | 56.90% | |||
| 2019-06-03 | Res_BUPT_2 | 55.34% | |||
| 2019-06-03 | Res_BUPT | 54.74% |