Why do these match? Explaining the behavior of image similarity models

Why do these match? Explaining the behavior of image similarity models Plummer, Bryan A.; Vasileva, Mariya I.; Petsiuk, Vitali; Saenko, Kate; Forsyth, David A. Vedaldi, Andrea; Bischof, Horst; Brox, Thomas; Frahm, Jan-Michael Explaining a deep learning model can help users understand its behavior and allow researchers to discern its shortcomings. Recent work has primarily focused on explaining models for tasks like image classiffication or visual question answering. In this paper, we introduce Salient Attributes for Network Explanation (SANE) to explain image similarity models, where a model's output is a score measuring the similarity of two inputs rather than a classiffication score. In this task, an explanation depends on both of the input images, so standard methods do not apply. Our SANE explanations pairs a saliency map identifying important image regions with an attribute that best explains the match. We find that our explanations provide additional information not typically captured by saliency maps alone, and can also improve performance on the classic task of attribute recognition. Our approach's ability to generalize is demonstrated on two datasets from diverse do- mains, Polyvore Outfits and Animals with Attributes 2. Code available at: https://github.com/VisionLearningGroup/SANE