EAPFusion: Intrinsic Evolving Auxiliary Prior Guidance for Infrared and Visible Image Fusion

TL;DR

EAPFusion introduces a scene-adaptive image fusion method using evolving intrinsic priors to dynamically generate convolutional kernels, improving infrared-visible fusion quality and downstream task performance.

Contribution

The paper proposes a novel scene-adaptive fusion approach with intrinsic priors that evolve across scales, replacing static models and enhancing fusion and downstream tasks.

Findings

Achieves state-of-the-art fusion results on multiple datasets.

Improves downstream semantic segmentation performance.

Demonstrates robustness in cross-dataset evaluations.

Abstract

Infrared-visible image fusion aims to create an information-rich fused image by integrating the complementary thermal saliency from infrared sensing and fine textures from visible imaging. Such accurate fusion is essential for real-world perception applications in complex scenes, including nighttime autonomous driving, search and rescue, and surveillance, and can further benefit downstream tasks such as semantic segmentation. However, most existing fusion methods rely upon static trained weights that cannot adapt to scene-specific content at inference time, and often suffer from a granularity mismatch when coarse auxiliary semantics are injected, which makes it difficult to simultaneously highlight targets and preserve details. In this work, we propose EAPFusion to address these issues by using self-evolving intrinsic priors instead of relying on external auxiliary models. Concretely, EAPFusion maintains a compact set of intrinsic priors and progressively updates them across scales. These evolved priors are utilized to dynamically generate convolutional kernels, shifting the paradigm from fixed, pre-trained filters to instance-adaptive parameters via prior-conditioned dynamic convolution. Furthermore, we design a channel-level fusion module that shuffles and interleaves infrared and visible channels, applying local channel mixing to boost cross-modal complementarity. Experiments on different datasets, including cross-dataset evaluation and semantic segmentation, show that the proposed method achieves state-of-the-art quantitative and qualitative fusion results, and consistently boosts downstream performance. Code is coming soon.