Consistency-constrained Non-negative Coding for Tracking

Xiaolin Tian (1) , Licheng Jiao (1) , Zhipeng Gan (1) , Chaohui Wang (2) , Xiaoli Zheng (1)

1 Key Laboratory of Intelligent Perception and Image Understanding
2 LIGM - Laboratoire d'Informatique Gaspard-Monge

A novel visual object tracking method based on consistency-constrained non-negative coding (CNC) is proposed in this paper. For the purpose of computational efficiency, superpixels are firstly extracted from each observed video frame. And then CNC is performed based on those obtained superpixels, where the locality on manifold is preserved by enforcing the temporal and spatial smoothness. The coding result is achieved via an iterative update scheme, which is proved to converge. The proposed method enhances the coding stability and makes the tracker more robust for object tracking. The tracking performance has been evaluated based on ten challenging benchmark sequences involving drastic motion, partial or severe occlusions, large variation in pose, and illumination variation. The experimental results demonstrate the superior performance of our method in comparison with ten state-of-art trackers.

Mots clés: Object tracking; non-negative matrix factorization; non-negative coding; classifier

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Xiaolin Tian, Licheng Jiao, Zhipeng Gan, Chaohui Wang, Xiaoli Zheng. Consistency-constrained Non-negative Coding for Tracking. IEEE Transactions on Circuits and Systems for Video Technology, 2017, 27 (4), pp.880-891. ⟨10.1109/TCSVT.2015.2501740⟩. ⟨hal-01624657⟩