Discriminative fusion of shape and appearance features for human pose estimation
作者：
Highlights：
• Discriminative fusion of shape and appearance features for 3D human pose estimation.
• We present a context-based appearance descriptor and fuse it with a shape descriptor.
• We propose a novel localized decision level fusion algorithm.
• We compare our fusion method with a feature level fusion and a kernel level fusion.
• Our fusion method improves the performance of 3D human pose estimation.
摘要
•Discriminative fusion of shape and appearance features for 3D human pose estimation.•We present a context-based appearance descriptor and fuse it with a shape descriptor.•We propose a novel localized decision level fusion algorithm.•We compare our fusion method with a feature level fusion and a kernel level fusion.•Our fusion method improves the performance of 3D human pose estimation.
论文关键词：3D human pose estimation,Discriminative fusion,Mixture of regressors,Appearance descriptors,Shape descriptors
论文评审过程：Received 22 October 2012, Revised 8 April 2013, Accepted 21 May 2013, Available online 6 June 2013.
论文官网地址：https://doi.org/10.1016/j.patcog.2013.05.019
