- Task 1 - Detection
- Task 2 - Detection-Linking
- Task 3 - Detection-Recognition
- Task 4 - Detection-Recognition-Linking
method: MapTest2024-05-05
Authors: Hongen Liu
Affiliation: Tianjin University
method: MapText Detection-Linking Strong Pipeline2024-05-06
Authors: Yu Xie, Ziyue Wang, Jielei Zhang
Affiliation: Bilibili Inc.
Description: In the Detection-linking task of MapText, we used ViTAE-v2 to extract global features, utilizing an encoder-decoder network architecture (DeepSolo). Data augmentation techniques such as cropping, scaling, saturation, and contrast adjustment were applied. Pre-training was conducted using available real datasets (TextOCR, TotalText, IC15, MLT2017). The model was fine-tuned on the MapText dataset, and post-processing methods were employed.
Zhang, Q., Xu, Y., Zhang, J., & Tao, D. (2023). Vitaev2: Vision transformer advanced by exploring inductive bias for image recognition and beyond. International Journal of Computer Vision, 131(5), 1141-1162.
Ye, M., Zhang, J., Zhao, S., Liu, J., Liu, T., Du, B., & Tao, D. (2023). Deepsolo: Let transformer decoder with explicit points solo for text spotting. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 19348-19357).
method: Baseline TESTR Checkpoint2024-03-26
Authors: Organizers
Affiliation: ICDAR'24 RRC-MapText
Description: TESTR checkpoint is used without any additional modifications or finetuning. The model checkpoint version with polygon prediction head and fine-tuned on TotalText was used. (Note that no links are predicted.)
| Date | Method | Quality | F-score | Tightness | Precision | Recall | |||
|---|---|---|---|---|---|---|---|---|---|
| 2024-05-05 | MapTest | 41.94% | 56.35% | 74.42% | 44.16% | 77.84% | |||
| 2024-05-06 | MapText Detection-Linking Strong Pipeline | 41.55% | 55.08% | 75.44% | 43.24% | 75.83% | |||
| 2024-03-26 | Baseline TESTR Checkpoint | 35.47% | 47.29% | 75.00% | 37.77% | 63.22% | |||
| 2024-03-26 | DS-LP | 35.04% | 50.25% | 69.73% | 39.39% | 69.37% |