A review on Person re-identification

Abstract

Person re-identification is an important task in intelligent video surveillance systems. Person reidentification refers to identifying the same person across a network of cameras with non-overlapping fields of view. It can be formally defined as establishing correspondence between images of a person taken from different camera. Person re-identification is a challenging task due to inter/intra illumination changes, occlusions, pose variations and subject intrinsic variations. Person re-identification is used to ensure security in park, shopping mall and universities. This paper gives a review of person re-identification techniques implemented using various datasets.