Visual Similarity Attention
This work addresses the lack of transparency in similarity models for computer vision, offering a new paradigm that enhances explainability and performance, though it is incremental in advancing existing methods.
The paper tackles the problem of explaining why images are similar or dissimilar in distance metric learning by proposing a gradient-based attention method for generating visual similarity explanations, and demonstrates that this approach improves model generalizability and explainability across tasks like image retrieval and person re-identification.
While there has been substantial progress in learning suitable distance metrics, these techniques in general lack transparency and decision reasoning, i.e., explaining why the input set of images is similar or dissimilar. In this work, we solve this key problem by proposing the first method to generate generic visual similarity explanations with gradient-based attention. We demonstrate that our technique is agnostic to the specific similarity model type, e.g., we show applicability to Siamese, triplet, and quadruplet models. Furthermore, we make our proposed similarity attention a principled part of the learning process, resulting in a new paradigm for learning similarity functions. We demonstrate that our learning mechanism results in more generalizable, as well as explainable, similarity models. Finally, we demonstrate the generality of our framework by means of experiments on a variety of tasks, including image retrieval, person re-identification, and low-shot semantic segmentation.