Summary of Daf-net: a Dual-branch Feature Decomposition Fusion Network with Domain Adaptive For Infrared and Visible Image Fusion, by Jian Xu et al.
DAF-Net: A Dual-Branch Feature Decomposition Fusion Network with Domain Adaptive for Infrared and Visible Image Fusion
by Jian Xu, Xin He
First submitted to arxiv on: 18 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel approach to infrared and visible image fusion, the dual-branch feature decomposition fusion network (DAF-Net), is proposed. This method combines complementary information from both modalities using a Restormer-based base encoder for global structural features and an Invertible Neural Network (INN) detail encoder for texture details. The Multi-Kernel Maximum Mean Discrepancy (MK-MMD) algorithm aligns the latent feature spaces of visible and infrared images, improving fused image quality. Experimental results demonstrate superior performance across multiple datasets, enhancing both visual quality and fusion performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Image fusion combines information from different modalities to create a more complete understanding of a scene. This paper proposes a new way to do this called DAF-Net. It uses two types of networks to capture different features: Restormer for global structure and Invertible Neural Networks (INN) for texture details. The network also includes an algorithm that helps the two modalities work together better. The results show that this method is better than others at creating fused images. |
Keywords
» Artificial intelligence » Encoder » Neural network
