Efficient and Accurate Scene Text Recognition with Cascaded-Transformers
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916660945879040 |
|---|---|
| author | Ozkan, Savas Maracani, Andrea Kim, Hyowon Cho, Sijun Noh, Eunchung Min, Jeongwon Cho, Jung Min Ozay, Mete |
| author_facet | Ozkan, Savas Maracani, Andrea Kim, Hyowon Cho, Sijun Noh, Eunchung Min, Jeongwon Cho, Jung Min Ozay, Mete |
| contents | In recent years, vision transformers with text decoder have demonstrated remarkable performance on Scene Text Recognition (STR) due to their ability to capture long-range dependencies and contextual relationships with high learning capacity. However, the computational and memory demands of these models are significant, limiting their deployment in resource-constrained applications. To address this challenge, we propose an efficient and accurate STR system. Specifically, we focus on improving the efficiency of encoder models by introducing a cascaded-transformers structure. This structure progressively reduces the vision token size during the encoding step, effectively eliminating redundant tokens and reducing computational cost. Our experimental results confirm that our STR system achieves comparable performance to state-of-the-art baselines while substantially decreasing computational requirements. In particular, for large-models, the accuracy remains same, 92.77 to 92.68, while computational complexity is almost halved with our structure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_18883 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Efficient and Accurate Scene Text Recognition with Cascaded-Transformers Ozkan, Savas Maracani, Andrea Kim, Hyowon Cho, Sijun Noh, Eunchung Min, Jeongwon Cho, Jung Min Ozay, Mete Computer Vision and Pattern Recognition In recent years, vision transformers with text decoder have demonstrated remarkable performance on Scene Text Recognition (STR) due to their ability to capture long-range dependencies and contextual relationships with high learning capacity. However, the computational and memory demands of these models are significant, limiting their deployment in resource-constrained applications. To address this challenge, we propose an efficient and accurate STR system. Specifically, we focus on improving the efficiency of encoder models by introducing a cascaded-transformers structure. This structure progressively reduces the vision token size during the encoding step, effectively eliminating redundant tokens and reducing computational cost. Our experimental results confirm that our STR system achieves comparable performance to state-of-the-art baselines while substantially decreasing computational requirements. In particular, for large-models, the accuracy remains same, 92.77 to 92.68, while computational complexity is almost halved with our structure. |
| title | Efficient and Accurate Scene Text Recognition with Cascaded-Transformers |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.18883 |