IPAD: Iterative, Parallel, and Diffusion-Based Network for Scene Text Recognition
International Journal of Computer Vision · 2025-08 · 133 (8) · 5589-5609
DOI: 10.1007/s11263-025-02443-1 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computer Vision, online erschienen am 14.05.2025. Verfasst von Xiaomeng Yang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 113 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Key Research Program of Frontier Science, Chinese Academy of Sciences.
Abstract (Originalsprache)
Nowadays, scene text recognition has attracted more and more attention due to its diverse applications. Most state-of-the-art methods adopt an encoder-decoder framework with the attention mechanism, autoregressively generating text from left to right. Despite the convincing performance, this sequential decoding strategy constrains the inference speed. Conversely, non-autoregressive models provide faster, simultaneous predictions but often sacrifice accuracy. Although utilizing an explicit language model can improve performance, it burdens the computational load. Besides, separating linguistic knowledge from vision information may harm the final prediction. In this paper, we propose an alternative solution that uses a parallel and iterative decoder that adopts an easy-first decoding strategy. Furthermore, we regard text recognition as an image-based conditional text generation task and utilize the discrete diffusion strategy, ensuring exhaustive exploration of bidirectional contextual information. Extensive experiments demonstrate that the proposed approach achieves superior results on the benchmark datasets, including both Chinese and English text images.
Förderung
- National Natural Science Foundation of China
- Key Research Program of Frontier Science, Chinese Academy of Sciences
Zitieren
APA
Yang, X., Qiao, Z., & Zhou, Y. (2025). IPAD: Iterative, Parallel, and Diffusion-Based Network for Scene Text Recognition. International Journal of Computer Vision, 133(8), 5589-5609. https://doi.org/10.1007/s11263-025-02443-1
Vancouver
Yang X, Qiao Z, Zhou Y. IPAD: Iterative, Parallel, and Diffusion-Based Network for Scene Text Recognition. International Journal of Computer Vision. 2025;133(8):5589-5609. doi:10.1007/s11263-025-02443-1
BibTeX
@article{Yang2025,
author = {Yang, Xiaomeng and Qiao, Zhi and Zhou, Yu},
title = {IPAD: Iterative, Parallel, and Diffusion-Based Network for Scene Text Recognition},
journal = {International Journal of Computer Vision},
year = {2025},
volume = {133},
number = {8},
pages = {5589--5609},
doi = {10.1007/s11263-025-02443-1}
}
EUVIMED