[Announced by Harbin Institute of Technology, Shenzhen (Fan Nana/Text) Recently, IEEE CVPR 2019, the world's top computer vision conference, announced the results of the paper selection. The paper "Target-Aware Deep Tracking" written by the Visual Perception Team of the School of Computer Science of Harbin Institute of Technology, Shenzhen was selected. The first author of the paper is Li Xin, a 2014 doctoral student, and the correspondent author is Professor He Zhenyu.

    The selected paper "Target-Aware Deep Tracking" of Harbin Institute of Technology, Shenzhen focuses to solving the validity of the generalized depth feature model in the field of visual tracking. In this paper, a novel visual target tracking framework is proposed, and the corresponding feature model is obtained by constructing the loss function suitable for the target tracking task. This model can effectively deal with the problems of target deformation, occlusion and background interference in target tracking, and solve the problem of lack of offline data annotation in visual object tracking. (Editor Wu Ruichan)