Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.03349 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908859448164352 |
|---|---|
| author | Coquenet, Denis |
| author_facet | Coquenet, Denis |
| contents | Recent advances in text recognition led to a paradigm shift for page-level recognition, from multi-step segmentation-based approaches to end-to-end attention-based ones. However, the naïve character-level autoregressive decoding process results in long prediction times: it requires several seconds to process a single page image on a modern GPU. We propose the Meta Document Attention Network (Meta-DAN) as a novel decoding strategy to reduce the prediction time while enabling a better context modeling. It relies on two main components: windowed queries, to process several transformer queries altogether, enlarging the context modeling with near future; and multi-token predictions, whose goal is to predict several tokens per query instead of only the next one. We evaluate the proposed approach on 10 full-page handwritten datasets and demonstrate state-of-the-art results on average in terms of character error rate. Source code and weights of trained models are available at https://github.com/FactoDeepLearning/meta_dan. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_03349 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Meta-DAN: towards an efficient prediction strategy for page-level handwritten text recognition Coquenet, Denis Computer Vision and Pattern Recognition Recent advances in text recognition led to a paradigm shift for page-level recognition, from multi-step segmentation-based approaches to end-to-end attention-based ones. However, the naïve character-level autoregressive decoding process results in long prediction times: it requires several seconds to process a single page image on a modern GPU. We propose the Meta Document Attention Network (Meta-DAN) as a novel decoding strategy to reduce the prediction time while enabling a better context modeling. It relies on two main components: windowed queries, to process several transformer queries altogether, enlarging the context modeling with near future; and multi-token predictions, whose goal is to predict several tokens per query instead of only the next one. We evaluate the proposed approach on 10 full-page handwritten datasets and demonstrate state-of-the-art results on average in terms of character error rate. Source code and weights of trained models are available at https://github.com/FactoDeepLearning/meta_dan. |
| title | Meta-DAN: towards an efficient prediction strategy for page-level handwritten text recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.03349 |