Semi-Supervised Multiple Instance Learning and its application in visual tracking
Yu Chen Zhou, Anlong Ming · 2016
In this paper, a novel Semi-Supervised Multiple Instance Learning (Semi-MIL) approach is presented. Compared with conventional approaches, we utilize a kind of “bag of instances” representation in the semi-supervised learning process, which provides an effective way to use the unlabeled data in multiple instance learning problem. We formulate the problem with a graph model based on the Minimax kernel. In addition, the Semi-MIL algorithm is readily applied for visual tracking, which can resolve the ambiguities during the tracking process. The presented approach is validated on several benchmark videos for visual tracking and MUSKs dataset for classification, the competitive experimental results demonstrate the effectiveness of our approach.