Discovery-and-Selection: Towards Optimal Multiple Instance Learning for Weakly Supervised Object Detection. (arXiv:2110.09060v2 [cs.CV] UPDATED)

Web: http://arxiv.org/abs/2110.09060

Weakly supervised object detection (WSOD) is a challenging task that requires
simultaneously learn object classifiers and estimate object locations under the
supervision of image category labels. A major line of WSOD methods roots in
multiple instance learning which regards images as bags of instances and
selects positive instances from each bag to learn the detector. However, a
grand challenge emerges when the detector inclines to converge to
discriminative parts of objects rather than the whole objects. In this paper,
under …

arxiv cv detection discovery learning