Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.02220 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912257703673856 |
|---|---|
| author | Shen, Zhihua Chen, Siyang Wang, Han Zhang, Tongsu Zhang, Xiaohu Xu, Xiangpeng Yang, Xia |
| author_facet | Shen, Zhihua Chen, Siyang Wang, Han Zhang, Tongsu Zhang, Xiaohu Xu, Xiangpeng Yang, Xia |
| contents | Multi-frame infrared small target detection (IRSTD) plays a crucial role in low-altitude and maritime surveillance. The hybrid architecture combining CNNs and Transformers shows great promise for enhancing multi-frame IRSTD performance. In this paper, we propose LVNet, a simple yet powerful hybrid architecture that redefines low-level feature learning in hybrid frameworks for multi-frame IRSTD. Our key insight is that the standard linear patch embeddings in Vision Transformers are insufficient for capturing the scale-sensitive local features critical to infrared small targets. To address this limitation, we introduce a multi-scale CNN frontend that explicitly models local features by leveraging the local spatial bias of convolution. Additionally, we design a U-shaped video Transformer for multi-frame spatiotemporal context modeling, effectively capturing the motion characteristics of targets. Experiments on the publicly available datasets IRDST and NUDT-MIRSDT demonstrate that LVNet outperforms existing state-of-the-art methods. Notably, compared to the current best-performing method, LMAFormer, LVNet achieves an improvement of 5.63\% / 18.36\% in nIoU, while using only 1/221 of the parameters and 1/92 / 1/21 of the computational cost. Ablation studies further validate the importance of low-level representation learning in hybrid architectures. Our code and trained models are available at https://github.com/ZhihuaShen/LVNet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_02220 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Low-Level Matters: An Efficient Hybrid Architecture for Robust Multi-frame Infrared Small Target Detection Shen, Zhihua Chen, Siyang Wang, Han Zhang, Tongsu Zhang, Xiaohu Xu, Xiangpeng Yang, Xia Computer Vision and Pattern Recognition Multi-frame infrared small target detection (IRSTD) plays a crucial role in low-altitude and maritime surveillance. The hybrid architecture combining CNNs and Transformers shows great promise for enhancing multi-frame IRSTD performance. In this paper, we propose LVNet, a simple yet powerful hybrid architecture that redefines low-level feature learning in hybrid frameworks for multi-frame IRSTD. Our key insight is that the standard linear patch embeddings in Vision Transformers are insufficient for capturing the scale-sensitive local features critical to infrared small targets. To address this limitation, we introduce a multi-scale CNN frontend that explicitly models local features by leveraging the local spatial bias of convolution. Additionally, we design a U-shaped video Transformer for multi-frame spatiotemporal context modeling, effectively capturing the motion characteristics of targets. Experiments on the publicly available datasets IRDST and NUDT-MIRSDT demonstrate that LVNet outperforms existing state-of-the-art methods. Notably, compared to the current best-performing method, LMAFormer, LVNet achieves an improvement of 5.63\% / 18.36\% in nIoU, while using only 1/221 of the parameters and 1/92 / 1/21 of the computational cost. Ablation studies further validate the importance of low-level representation learning in hybrid architectures. Our code and trained models are available at https://github.com/ZhihuaShen/LVNet. |
| title | Low-Level Matters: An Efficient Hybrid Architecture for Robust Multi-frame Infrared Small Target Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.02220 |