Tracklet and Signature Representation for Multi-shot Person Re-Identification. - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2018

Tracklet and Signature Representation for Multi-shot Person Re-Identification.

Abstract

Video surveillance has become more and more important in many domains for their security and safety. Person Re-Identification (Re-ID) is one of the most interesting subjects in this area. The Re-ID system is divided into two main stages: i) extracting feature representations to construct a person’s appearance sig- nature and ii) establishing the correspondence/matching by learning similarity metrics or ranking functions. However, appearance based person Re-ID is a challenging task due to similarity of human’s appearance and visual ambiguities across different cameras. This paper provides a representation of the appearance descriptors, called signatures, for multi-shot Re-ID. First, we will present the tracklets, i.e trajectories of persons. Then, we compute the signature and represent it based on the approach of Part Appearance Mixture (PAM). An evaluation of the quality of this signature representation is also described in order to essentially solve the problems of high variance in a person’s appearance, occlusions, illumination changes and person’s orientation/pose. To deal with variance in a person’s appearance, we represent it as a set of multi-modal feature distributions modeled by Gaussian Mixture Model (GMM). Experiments and results on two public datasets and on our own dataset show good performance.
Fichier principal
Vignette du fichier
SSD paper Salwa.pdf (953.02 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01849457 , version 1 (26-07-2018)
hal-01849457 , version 2 (13-08-2018)

Identifiers

  • HAL Id : hal-01849457 , version 2

Cite

Salwa Baabou, Furqan M Khan, François Bremond, Awatef Ben Frad, Mohamed Amine Farah, et al.. Tracklet and Signature Representation for Multi-shot Person Re-Identification. . SSD 2018 - International Multi-Conference on Systems, Signals and Devices, Mar 2018, Hammamet, Tunisia. pp.1-6. ⟨hal-01849457v2⟩
164 View
199 Download

Share

Gmail Facebook X LinkedIn More