Research map / Vision-language and multimodal / Human and embodied modeling
Bridging Human Evaluation to Infrared and Visible Image Fusion
Jinyuan Liu, Xingyuan Li, Qingyun Mei, Haoyuan Xu, Zhiying Jiang, Long Ma, Risheng Liu, Xin Fan
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 reward-model score branch (ablation) | EVAFusion · M3FD | PSNR: 64.81 -> 65.09 | Improved |
| add reward-model heatmap branch (ablation) | EVAFusion · M3FD | PSNR: 64.93 -> 65.09 | Improved |
| add SAM region segmentation (ablation) | EVAFusion · M3FD | PSNR: 63.02 -> 65.09 | Improved |
| add SAM region segmentation (ablation) | EVAFusion · TNO | CC: 0.48 -> 0.51 | Improved |
| add reward-model score branch (ablation) | EVAFusion · TNO | CC: 0.50 -> 0.51 | No clear effect |
| use GRPO with regional advantages instead of DPO (ablation) | EVAFusion · TNO | PSNR: 63.98 -> 65.43 | Improved |
| use GRPO instead of PPO (ablation) | EVAFusion · TNO | PSNR: 63.59 -> 65.43 | Improved |
| fine-tune with PPO versus baseline (ablation) | DCEvo · TNO | PSNR: 63.83 -> 63.59 | Hurt |
| fine-tune with DPO versus baseline (ablation) | DCEvo · RoadScene | PSNR: 59.66 -> 61.50 | Improved |
| add RLHF fine-tuning with human-feedback reward | DCEvo · RoadScene | PSNR: 59.66 -> 61.84 | Improved |
| add RLHF fine-tuning with human-feedback reward | DCEvo · M3FD detection | mAP: 58.23 -> 62.23 | Improved |
| add RLHF fine-tuning with human-feedback reward | DCEvo · TNO | NIQE: 6.25 -> 5.37 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DCEvo — DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion (CVPR 2025) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Revisiting Generative Infrared and Visible Image Fusion Based on Human Cognitive Laws (NeurIPS 2025 · similarity 0.87)
- DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion (CVPR 2025 · similarity 0.85)
- Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark (NeurIPS 2025 · similarity 0.85)
- Every SAM Drop Counts: Embracing Semantic Priors for Multi-Modality Image Fusion and Beyond (CVPR 2025 · similarity 0.85)
- Evaluation of Visual Processing Should Be Human-Centered, Not Metric-Centered (NeurIPS 2026 · similarity 0.84)
- ControlFusion: A Controllable Image Fusion Network with Language-Vision Degradation Prompts (NeurIPS 2025 · similarity 0.84)
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.
