SAMSOD: Rethinking SAM Optimization for RGB-T Salient Object Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918154437918720 |
|---|---|
| author | Liu, Zhengyi Wang, Xinrui Fang, Xianyong Tu, Zhengzheng Wang, Linbo |
| author_facet | Liu, Zhengyi Wang, Xinrui Fang, Xianyong Tu, Zhengzheng Wang, Linbo |
| contents | RGB-T salient object detection (SOD) aims to segment attractive objects by combining RGB and thermal infrared images. To enhance performance, the Segment Anything Model has been fine-tuned for this task. However, the imbalance convergence of two modalities and significant gradient difference between high- and low- activations are ignored, thereby leaving room for further performance enhancement. In this paper, we propose a model called \textit{SAMSOD}, which utilizes unimodal supervision to enhance the learning of non-dominant modality and employs gradient deconfliction to reduce the impact of conflicting gradients on model convergence. The method also leverages two decoupled adapters to separately mask high- and low-activation neurons, emphasizing foreground objects by enhancing background learning. Fundamental experiments on RGB-T SOD benchmark datasets and generalizability experiments on scribble supervised RGB-T SOD, fully supervised RGB-D SOD datasets and full-supervised RGB-D rail surface defect detection all demonstrate the effectiveness of our proposed method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_03689 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SAMSOD: Rethinking SAM Optimization for RGB-T Salient Object Detection Liu, Zhengyi Wang, Xinrui Fang, Xianyong Tu, Zhengzheng Wang, Linbo Computer Vision and Pattern Recognition RGB-T salient object detection (SOD) aims to segment attractive objects by combining RGB and thermal infrared images. To enhance performance, the Segment Anything Model has been fine-tuned for this task. However, the imbalance convergence of two modalities and significant gradient difference between high- and low- activations are ignored, thereby leaving room for further performance enhancement. In this paper, we propose a model called \textit{SAMSOD}, which utilizes unimodal supervision to enhance the learning of non-dominant modality and employs gradient deconfliction to reduce the impact of conflicting gradients on model convergence. The method also leverages two decoupled adapters to separately mask high- and low-activation neurons, emphasizing foreground objects by enhancing background learning. Fundamental experiments on RGB-T SOD benchmark datasets and generalizability experiments on scribble supervised RGB-T SOD, fully supervised RGB-D SOD datasets and full-supervised RGB-D rail surface defect detection all demonstrate the effectiveness of our proposed method. |
| title | SAMSOD: Rethinking SAM Optimization for RGB-T Salient Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.03689 |