Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2510.12605 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914284676579328 |
|---|---|
| author | Li, Runting Lian, Shijie Li, Hua Li, Yutong Wu, Wenhui Kwong, Sam |
| author_facet | Li, Runting Lian, Shijie Li, Hua Li, Yutong Wu, Wenhui Kwong, Sam |
| contents | Underwater Salient Object Detection (USOD) faces significant challenges, including underwater image quality degradation and domain gaps. Existing methods tend to ignore the physical principles of underwater imaging or simply treat degradation phenomena in underwater images as interference factors that must be eliminated, failing to fully exploit the valuable information they contain. We propose WaterFlow, a rectified flow-based framework for underwater salient object detection that innovatively incorporates underwater physical imaging information as explicit priors directly into the network training process and introduces temporal dimension modeling, significantly enhancing the model's capability for salient object identification. On the USOD10K dataset, WaterFlow achieves a 0.072 gain in S_m, demonstrating the effectiveness and superiority of our method. https://github.com/Theo-polis/WaterFlow. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_12605 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WaterFlow: Explicit Physics-Prior Rectified Flow for Underwater Saliency Mask Generation Li, Runting Lian, Shijie Li, Hua Li, Yutong Wu, Wenhui Kwong, Sam Computer Vision and Pattern Recognition Underwater Salient Object Detection (USOD) faces significant challenges, including underwater image quality degradation and domain gaps. Existing methods tend to ignore the physical principles of underwater imaging or simply treat degradation phenomena in underwater images as interference factors that must be eliminated, failing to fully exploit the valuable information they contain. We propose WaterFlow, a rectified flow-based framework for underwater salient object detection that innovatively incorporates underwater physical imaging information as explicit priors directly into the network training process and introduces temporal dimension modeling, significantly enhancing the model's capability for salient object identification. On the USOD10K dataset, WaterFlow achieves a 0.072 gain in S_m, demonstrating the effectiveness and superiority of our method. https://github.com/Theo-polis/WaterFlow. |
| title | WaterFlow: Explicit Physics-Prior Rectified Flow for Underwater Saliency Mask Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.12605 |