Features analysis of person re-identification techniques

Person re-identification across different surveillance cameras has become one of the recent and challenging task in the area of intelligent video surveillance system. Although many methods have been introduced for re-identification, certain issues like illumination, pose variation and occlusion are unsolved. In all the existing methods, features are extracted from the image using various learning methods. The aim of this work is to analyze the importance of feature for re-identification. The colour, texture and shape features has taken for the analysis. The database images are collected from multiple benchmark databases. The analysis shows that every feature has some unique properties and works better for some issues.