Visual object tracking via a manifold regularized discriminative dual dictionary model
作者：
Highlights：
• We propose a Manifold Regularized Discriminative Dual Dictionary (MD3) model.
• MD3 model introduces both spatial and temporal information, which are represented by discriminative dictionary and manifold regularization, respectively.
• An alternating optimization method is introduced to solve MD3 model.
• Particle Filtering tracking framework is utilized to incorporate MD3 model, in which observation model is represented by reconstruction error.
摘要
•We propose a Manifold Regularized Discriminative Dual Dictionary (MD3) model.•MD3 model introduces both spatial and temporal information, which are represented by discriminative dictionary and manifold regularization, respectively.•An alternating optimization method is introduced to solve MD3 model.•Particle Filtering tracking framework is utilized to incorporate MD3 model, in which observation model is represented by reconstruction error.
论文关键词：Visual tracking,Dictionary learning,Dual dictionary,Manifold regularization
论文评审过程：Received 21 June 2016, Revised 30 January 2019, Accepted 10 February 2019, Available online 28 February 2019, Version of Record 5 March 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.02.008
