学术速递｜2026年4月 arXiv 图像融合论文汇总（04.01–04.30）10 篇最新成果

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。

---

1. Thermal background reduction for mid-infrared imaging by low-rank background and sparse point-source modelling

https://arxiv.org/abs/2604.22351v1

---

2. The First Challenge on Remote Sensing Infrared Image Super-Resolution at NTIRE 2026: Benchmark Results and Method Overview

https://arxiv.org/abs/2604.21312v1

---

3. CoFusion: Multispectral and Hyperspectral Image Fusion via Spectral Coordinate Attention

https://arxiv.org/abs/2604.10584v2

---

4. Dual-Branch Remote Sensing Infrared Image Super-Resolution

https://arxiv.org/abs/2604.10112v2

---

5. 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/2604.09030v1

---

6. 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/2604.08924v1

---

7. 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/2604.08922v1

---

8. ASSR-Net: Anisotropic Structure-Aware and Spectrally Recalibrated Network for Hyperspectral Image Fusion

https://arxiv.org/abs/2604.05742v1

---

9. EvaNet: Towards More Efficient and Consistent Infrared and Visible Image Fusion Assessment

https://arxiv.org/abs/2604.02896v1

---

10. Harmonized Tabular-Image Fusion via Gradient-Aligned Alternating Learning

https://arxiv.org/abs/2604.01579v1

---

往期推荐

---