This repository presents a couple of approaches to the problem of multi-view image classification. I faced this challenge during a hackathon in which I participated, and decided to share my code here. I've also written a Medium article to provide further details and explanations. Feel free to check it out !
This is great repo. Though, I wonder how can i use GRAD CAM to see the activated regions in two images. Also, instead of max pooling the stacked features, I am just concating them together to increase the number of feature length and connected to fully connected layer before the classification as it was giving slighlty better results