SSDFusion: A scene-semantic decomposition approach for visible and infrared image fusion

Rui Ming, Yixian Xiao, Xinyu Liu, Guolong Zheng, Guobao Xiao

Abstract: Highlights•Devises a scene-semantic feature decomposition and fusion strategy for image fusion.•Proposes a multi-level encoder-fusion network to mimic the downstream backbone.•Connects feature decomposition-based image fusion with high-level vision tasks.•Experiments have demonstrated the superiority of our proposed method.