检索说明以关键词【Image Fusion\ Infrared-Visible\ Infrared Image】在arXiv网站的文章标题中进行搜索并将时间限定为04.01-04.30共计检索10篇相关论文。包含CVPR*2、CVPRW*1、TPAMI*1。从整体结果来看当前研究主要围绕红外–可见光融合、多光谱/高光谱融合以及融合与图像恢复的联合建模展开尤其是在复杂退化场景噪声、模糊、低分辨率下的鲁棒融合问题受到广泛关注。同时部分工作开始引入扩散模型、多任务学习以及闭环优化机制推动融合方法从传统静态设计向更加灵活、自适应的方向发展。从发展趋势上看图像融合正逐步从单一任务建模走向“融合恢复感知”的统一框架强调多模态信息的协同优化与端到端建模能力与此同时生成模型如Diffusion与数据驱动方法的引入使得融合过程具备更强的表达能力与泛化潜力。未来融合领域有望在全退化统一建模、动态自适应融合策略以及评价体系标准化等方向持续深入并进一步与下游智能感知任务紧密结合提升实际应用价值。Thermal background reduction for mid-infrared imaging by low-rank background and sparse point-source modellinghttps://arxiv.org/abs/2604.22351v12. The First Challenge on Remote Sensing Infrared Image Super-Resolution at NTIRE 2026: Benchmark Results and Method Overviewhttps://arxiv.org/abs/2604.21312v13. CoFusion: Multispectral and Hyperspectral Image Fusion via Spectral Coordinate Attentionhttps://arxiv.org/abs/2604.10584v24. Dual-Branch Remote Sensing Infrared Image Super-Resolutionhttps://arxiv.org/abs/2604.10112v25. NTIRE 2026 The 3rd Restore Any Image Model (RAIM) Challenge: Multi-Exposure Image Fusion in Dynamic Scenes (Track 2) (已解读点击跳转)https://arxiv.org/abs/2604.09030v16. Customized Fusion: A Closed-Loop Dynamic Network for Adaptive Multi-Task-Aware Infrared-Visible Image Fusion (已解读点击跳转)https://arxiv.org/abs/2604.08924v17. Degradation-Robust Fusion: An Efficient Degradation-Aware Diffusion Framework for Multimodal Image Fusion in Arbitrary Degradation Scenarios (已解读点击跳转https://arxiv.org/abs/2604.08922v18. ASSR-Net: Anisotropic Structure-Aware and Spectrally Recalibrated Network for Hyperspectral Image Fusionhttps://arxiv.org/abs/2604.05742v19. EvaNet: Towards More Efficient and Consistent Infrared and Visible Image Fusion Assessmenthttps://arxiv.org/abs/2604.02896v110. Harmonized Tabular-Image Fusion via Gradient-Aligned Alternating Learninghttps://arxiv.org/abs/2604.01579v1往期推荐