Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.16364 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909933133365248 |
|---|---|
| author | Shih, Meng-Cheng Huang, Tsai-Ling Shih, Yu-Heng Shuai, Hong-Han Liu, Hsuan-Tung Yeh, Yi-Ren Huang, Ching-Chun |
| author_facet | Shih, Meng-Cheng Huang, Tsai-Ling Shih, Yu-Heng Shuai, Hong-Han Liu, Hsuan-Tung Yeh, Yi-Ren Huang, Ching-Chun |
| contents | Offline signature verification (OSV) is a frequently utilized technology in forensics. This paper proposes a new model, DetailSemNet, for OSV. Unlike previous methods that rely on holistic features for pair comparisons, our approach underscores the significance of fine-grained differences for robust OSV. We propose to match local structures between two signature images, significantly boosting verification accuracy. Furthermore, we observe that without specific architectural modifications, transformer-based backbones might naturally obscure local details, adversely impacting OSV performance. To address this, we introduce a Detail Semantics Integrator, leveraging feature disentanglement and re-entanglement. This integrator is specifically designed to enhance intricate details while simultaneously expanding discriminative semantics, thereby augmenting the efficacy of local structural matching. We evaluate our method against leading benchmarks in offline signature verification. Our model consistently outperforms recent methods, achieving state-of-the-art results with clear margins. The emphasis on local structure matching not only improves performance but also enhances the model's interpretability, supporting our findings. Additionally, our model demonstrates remarkable generalization capabilities in cross-dataset testing scenarios. The combination of generalizability and interpretability significantly bolsters the potential of DetailSemNet for real-world applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_16364 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DetailSemNet: Elevating Signature Verification through Detail-Semantic Integration Shih, Meng-Cheng Huang, Tsai-Ling Shih, Yu-Heng Shuai, Hong-Han Liu, Hsuan-Tung Yeh, Yi-Ren Huang, Ching-Chun Computer Vision and Pattern Recognition Offline signature verification (OSV) is a frequently utilized technology in forensics. This paper proposes a new model, DetailSemNet, for OSV. Unlike previous methods that rely on holistic features for pair comparisons, our approach underscores the significance of fine-grained differences for robust OSV. We propose to match local structures between two signature images, significantly boosting verification accuracy. Furthermore, we observe that without specific architectural modifications, transformer-based backbones might naturally obscure local details, adversely impacting OSV performance. To address this, we introduce a Detail Semantics Integrator, leveraging feature disentanglement and re-entanglement. This integrator is specifically designed to enhance intricate details while simultaneously expanding discriminative semantics, thereby augmenting the efficacy of local structural matching. We evaluate our method against leading benchmarks in offline signature verification. Our model consistently outperforms recent methods, achieving state-of-the-art results with clear margins. The emphasis on local structure matching not only improves performance but also enhances the model's interpretability, supporting our findings. Additionally, our model demonstrates remarkable generalization capabilities in cross-dataset testing scenarios. The combination of generalizability and interpretability significantly bolsters the potential of DetailSemNet for real-world applications. |
| title | DetailSemNet: Elevating Signature Verification through Detail-Semantic Integration |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.16364 |