Attend Refine Repeat: Active Box Proposal Generation via In-Out Localization

Abstract

The problem of computing category agnostic bounding box proposals is utilized as a core component in many computer vision tasks and thus has lately attracted a lot of attention. In this work we propose a new approach to tackle this problem that is based on an active strategy for generating box proposals that starts from a set of seed boxes, which are uniformly distributed on the image, and then progressively moves its attention on the promising image areas where it is more likely to discover well localized bounding box proposals. We call our approach AttractioNet and a core component of it is a CNN-based category agnostic object location refinement module that is capable of yielding accurate and robust bounding box predictions regardless of the object category. We extensively evaluate our AttractioNet approach on the COCO 2014 validation set as well as on the PASCAL VOC2007 test set, reporting for both of them state-of-the-art results that surpass the previous work in the field by a significant margin. Finally, we provide strong empirical evidence that our approach is capable to generalize to unseen categories. Project page:: https://github.com/gidariss/AttractioNet.