Salient feature based graph matching for person re-identification
作者：
Highlights：
• Computational symmetry and structure modeling for people re-identification.
• A new feature detector and descriptor based on ASIFT enriched by local symmetries.
• ASIFT locations properly selected in agreement with the distance from the symmetry axis.
• A new graph representation to catch structural relations.
• Results and comparisons with state-of-the-art methods experienced on the i-LIDS MCTS dataset.
摘要
Highlights•Computational symmetry and structure modeling for people re-identification.•A new feature detector and descriptor based on ASIFT enriched by local symmetries.•ASIFT locations properly selected in agreement with the distance from the symmetry axis.•A new graph representation to catch structural relations.•Results and comparisons with state-of-the-art methods experienced on the i-LIDS MCTS dataset.
论文关键词：Computational symmetry,Salient features,Graph representation and matching,People re-identification
论文评审过程：Received 2 January 2014, Revised 12 September 2014, Accepted 15 September 2014, Available online 4 November 2014.
论文官网地址：https://doi.org/10.1016/j.patcog.2014.09.011
