Compare and Contrast: Learning Prominent Visual Differences

Abstract

Relative attribute models can compare images in terms of all detected properties or attributes, exhaustively predicting which image is fancier, more natural, and so on without any regard to ordering. However, when humans compare images, certain differences will naturally stick out and come to mind first. These most noticeable differences, or prominent differences, are likely to be described first. In addition, many differences, although present, may not be mentioned at all. In this work, we introduce and model prominent differences, a rich new functionality for comparing images. We collect instance-level annotations of most noticeable differences, and build a model trained on relative attribute features that predicts prominent differences for unseen pairs. We test our model on the challenging UT-Zap50K shoes and LFW10 faces datasets, and outperform an array of baseline methods. We then demonstrate how our prominence model improves two vision tasks, image search and description generation, enabling more natural communication between people and vision systems.

Learning to Predict Visual Attributes in the Wild

“…Some of the earliest work related to attribute learning stem from a desire to learn to describe objects rather than predicting their identities [18,17,6,19]. Since then, extensive work has sought to explore several aspects of object attributes, including attribute-based zero-shot object classification [43,32,2], relative attribute comparison [52,62,10], and image search [61,41]. While research in compositional zero shot learning [71,57,51,31] also tackle object attributes, they target transformation of 'states' of objects, treat each instance as having only one state, and focus on predicting unseen compositions rather than the prediction of complete set of attributes for each object instance.…”