CVLGIVDec 6, 2020

FuseVis: Interpreting neural networks for image fusion using per-pixel saliency visualization

arXiv:2012.08932v111 citations
AI Analysis

This work addresses the challenge of interpreting unsupervised CNNs for image fusion, which is crucial for assessing their reliability in applications like medical imaging, where ground truth is unavailable.

This paper introduces FuseVis, a real-time visualization tool for interpreting neural networks used in image fusion. It computes per-pixel saliency maps to show the influence of input image pixels on the fused image, demonstrating that some fusion methods are better suited for specific clinical applications.

Image fusion helps in merging two or more images to construct a more informative single fused image. Recently, unsupervised learning based convolutional neural networks (CNN) have been utilized for different types of image fusion tasks such as medical image fusion, infrared-visible image fusion for autonomous driving as well as multi-focus and multi-exposure image fusion for satellite imagery. However, it is challenging to analyze the reliability of these CNNs for the image fusion tasks since no groundtruth is available. This led to the use of a wide variety of model architectures and optimization functions yielding quite different fusion results. Additionally, due to the highly opaque nature of such neural networks, it is difficult to explain the internal mechanics behind its fusion results. To overcome these challenges, we present a novel real-time visualization tool, named FuseVis, with which the end-user can compute per-pixel saliency maps that examine the influence of the input image pixels on each pixel of the fused image. We trained several image fusion based CNNs on medical image pairs and then using our FuseVis tool, we performed case studies on a specific clinical application by interpreting the saliency maps from each of the fusion methods. We specifically visualized the relative influence of each input image on the predictions of the fused image and showed that some of the evaluated image fusion methods are better suited for the specific clinical application. To the best of our knowledge, currently, there is no approach for visual analysis of neural networks for image fusion. Therefore, this work opens up a new research direction to improve the interpretability of deep fusion networks. The FuseVis tool can also be adapted in other deep neural network based image processing applications to make them interpretable.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes