Research map / Detection and segmentation / CNN features
RA-SOD: Reliability-Aware RGB-T Salient Object Detection under Modality Degradation
Hongbo Gao, Zhengyu Li, XUERU NIE, Dihao Zhu, lijun zhao, Yunke Wang, Chang Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add reliability-aware framework over ConTriNet baseline | RA-SOD · VT5000 | weighted F-measure: 0.846 -> 0.855 | Improved |
| use RA-SOD instead of ConTriNet on challenge set | RA-SOD · VT-IMAG | F_beta: 0.745 -> 0.765 | Improved |
| use RA-SOD instead of ConTriNet under synthetic degradation | RA-SOD · VT1000 synthetic degradations | F_beta^w: 0.895 -> 0.907 | Improved |
| use RA-SOD with Swin-B backbone | RA-SOD · VT5000 (Swin-B) | S_m: 0.923 -> 0.930 | Improved |
| add modality-specific adapter (ablation) | RCBR · VT5000 | F_beta^w: 0.851 -> 0.853 | Improved |
| add shared expert bank only (ablation) | RCBR · VT5000 | F_beta^w: 0.851 -> 0.851 | No clear effect |
| add both adapter and expert bank (ablation) | RCBR · VT5000 | F_beta^w: 0.851 -> 0.855 | Improved |
| add uncertainty-guided dual-stream decoding (ablation) | RA-SOD · VT1000 | F_beta^w: 0.908 -> 0.913 | Improved |
| add uncertainty-guided dual-stream decoding (ablation) | RA-SOD · VT-IMAG | F_beta^w: 0.728 -> 0.735 | Improved |
| add uncertainty guidance to dual-stream decoding (ablation) | RA-SOD · VT1000 degraded | F_beta^w: 0.903 -> 0.907 | Improved |
| use 3 experts instead of 1 (ablation) | RCBR · VT5000 | F_beta^w: 0.853 -> 0.855 | Improved |
| use 5 experts instead of 3 (ablation) | RCBR · VT5000 | F_beta^w: 0.855 -> 0.850 | Hurt |
Closest papers by meaning
- Uncertainty-Aware Modality Fusion for Unaligned RGB-T Salient Object Detection (CVPR 2026 · similarity 0.93)
- Efficient RGB-T Object Detection via Sparse Cross-Modality Fusion (ECCV 2026 · similarity 0.91)
- Divide-and-Conquer: Confluent Triple-Flow Network for RGB-T Salient Object Detection (TPAMI 2024 · similarity 0.91)
- Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network (AAAI 2025 · similarity 0.91)
- RGB-D Salient Object Detection with Cross-Modality Modulation and Selection (ECCV 2020 · similarity 0.91)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
