Abstract

Attributes possess appealing properties and benefit many computer vision problems, such as object recog-nition, learning with humans in the loop, and image retrieval. Whereas the existing work mainly pursues utiliz-ing attributes for various computer vision problems, we contend that the most basic problem—how to accuratelyand robustly detect attributes from images—has been left under explored. Especially, the existing work rarely explicitly tackles the need that attribute detectors should generalize well across different categories, includingthose previously unseen. Noting that this is analogousto the objective of multi-source domain generalization, if we treat each category as a domain, we provide a novel perspective to attribute detection and propose to gear the techniques in multi-source domain generalization for the purpose of learning cross-category generalizable attribute detectors. We validate our understanding and approach with extensive experiments on four challenging datasets and three different problems.