Research map / Detection and segmentation / CNN features
Efficient RGB-T Object Detection via Sparse Cross-Modality Fusion
CHAO TIAN, Zikun Zhou, Chao Yang, guoqing zhu, Zhenyu He
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 differential enhancement to group-wise fusion (ablation) | FER fusion · M3FD | mAP: 58.4 -> 58.6 | Improved |
| add group-wise fusion to differential enhancement (ablation) | FER fusion · M3FD | mAP: 58.1 -> 58.6 | Improved |
| add auxiliary backbone features (ablation) | FER · M3FD | AP50: 87.2 -> 88.5 | Improved |
| take auxiliary features from backbone instead of RPN features (ablation) | FER · M3FD | AP50: 88.1 -> 88.5 | Improved |
| add second refinement step (ablation) | FER · M3FD | AP50: 88.5 -> 89.8 | Improved |
| use rolling convolution instead of Conv2d (ablation) | FER · M3FD | AP50: 88.9 -> 89.8 | Improved |
| add denoising training (ablation) | FER · M3FD | AP50: 88.0 -> 89.8 | Improved |
| use rolling convolution instead of Shuffle (G=8) (ablation) | FER · M3FD | AP50: 88.9 -> 89.8 | Improved |
| use RoI size 5 instead of 3 (ablation) | FER · M3FD | AP50: 88.7 -> 89.8 | Improved |
| use RoI size 7 instead of 5 (ablation) | FER · M3FD | AP50: 89.8 -> 88.9 | Hurt |
| use 16 rolling groups instead of 8 (ablation) | FER · M3FD | AP50: 89.8 -> 88.9 | Hurt |
| sparse fusion instead of dense FPN fusion with DEGF (ablation) | SFEDet · M3FD | AP50: 88.4 -> 89.8 | Improved |
Closest papers by meaning
- RA-SOD: Reliability-Aware RGB-T Salient Object Detection under Modality Degradation (ECCV 2026 · similarity 0.91)
- Hamiltonian Asymmetric Fusion: One-Way Safe Directed Refinement under Modality Imbalance (ICML 2026 · similarity 0.89)
- Distribution-Aligned Multimodal Fusion for Robust Object Detection (CVPR 2026 · similarity 0.89)
- Uncertainty-Aware Modality Fusion for Unaligned RGB-T Salient Object Detection (CVPR 2026 · similarity 0.88)
- Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network (AAAI 2025 · similarity 0.87)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.87)
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.
