method: TencentOCR2023-03-21
Authors: Fan Yang, Lifu Wang, Huiwen Shi, Sicong Liu, Qingxiang Lin, Yuxin Wang，Haoxi Li, Weida Chen, Yushuo Guan, Minhui Wu, Chunchao Guo, Hongfa Wang, Wei Liu
Affiliation: TencentOCR
Description: We integrated the detection results of DBNet and Cascade MaskRCNN built with multiple Backbone architectures, combined with the Parseq English recognition model for recognition, and further improved the end-to-end tracking with ByteTrack. As a result, we obtained end-to-end tracking and trajectory recognition results.
method: LOGO2024-05-30
Authors: Hongen Liu, Di Sun, Jiahao Wang, Yi Liu, Gang Pan
Affiliation: College of Intelligence and Computing, Tianjin University；Tianjin University of Science and Technology; Baidu Inc.
Description: We propose a Language Collaboration and Glyph Perception Model, termed LOGO to enhance the performance of conventional text spotters through the integration of a synergy module. To achieve this goal, a language synergy classifier (LSC) is designed to explicitly discern text instances from background noise in the recognition stage. Besides, the glyph supervision and visual position mixture module are proposed to enhance the recognition accuracy of noisy text regions, and acquire more discriminative tracking features, respectively.
method: TransDeTR+HRNet2023-03-20
Authors: Tianyu Zhang, Qing Yang, Yangchun Wan, Yifan Sun, Zhigang Chang, Hang Su, Zhenfeng Su
Affiliation: AI Lab, Du Xiaoman Financial
Description: We use the baseline code TransDeTR and the ResNet backbone is replaced with HRNet, which outputs high-resolution feature maps.
| Date | Method | MOTA | MOTP | IDF1 | Mostly Matched | Partially Matched | Mostly Lost | |||
|---|---|---|---|---|---|---|---|---|---|---|
| 2023-03-21 | TencentOCR | 62.56% | 79.88% | 75.87% | 8114 | 1800 | 2663 | |||
| 2024-05-30 | LOGO | 51.36% | 77.57% | 65.70% | 5743 | 2100 | 4734 | |||
| 2023-03-20 | TransDeTR+HRNet | 43.52% | 78.15% | 62.27% | 4980 | 2264 | 5333 | |||
| 2024-03-25 | ldswo | 37.38% | 76.08% | 53.24% | 3324 | 2649 | 6604 | |||
| 2023-03-20 | Video Text Tracking for Dense and Small Text Based on PP-YOLOE-R and Sort Algorithm | 36.87% | 79.24% | 48.99% | 2123 | 3625 | 6829 | |||
| 2023-03-19 | TextTrack | 25.75% | 74.03% | 50.22% | 3302 | 2806 | 6469 | |||
| 2023-03-19 | SCUT-MMOCR-KS | 13.83% | 75.75% | 58.41% | 6924 | 2622 | 3031 |