DIDFuse: Deep Image Decomposition for Infrared and Visible Image Fusion
This addresses the problem of combining infrared and visible images for enhanced visual analysis, such as in surveillance or remote sensing, but it is incremental as it builds on existing auto-encoder methods with a specific decomposition approach.
The paper tackles infrared and visible image fusion by proposing an auto-encoder network that decomposes images into background and detail features, merging them to produce fused images with highlighted targets and rich textures, achieving state-of-the-art results in qualitative and quantitative evaluations.
Infrared and visible image fusion, a hot topic in the field of image processing, aims at obtaining fused images keeping the advantages of source images. This paper proposes a novel auto-encoder (AE) based fusion network. The core idea is that the encoder decomposes an image into background and detail feature maps with low- and high-frequency information, respectively, and that the decoder recovers the original image. To this end, the loss function makes the background/detail feature maps of source images similar/dissimilar. In the test phase, background and detail feature maps are respectively merged via a fusion module, and the fused image is recovered by the decoder. Qualitative and quantitative results illustrate that our method can generate fusion images containing highlighted targets and abundant detail texture information with strong robustness and meanwhile surpass state-of-the-art (SOTA) approaches.