Lavoisier S.A.S.
14 rue de Provigny
94236 Cachan cedex

Heures d'ouverture 08h30-12h30/13h30-17h30
Tél.: +33 (0)1 47 40 67 00
Fax: +33 (0)1 47 40 67 02

Url canonique :
Url courte ou permalien :

Human Re-Identification, Softcover reprint of the original 1st ed. 2016 Multimedia Systems and Applications Series

Langue : Anglais

Auteur :

Couverture de l’ouvrage Human Re-Identification
This book covers aspects of human re-identification problems related to computer vision and machine learning. Working from a practical perspective, it introduces novel algorithms and designs for human re-identification that bridge the gap between research and reality. The primary focus is on building a robust, reliable, distributed and scalable smart surveillance system that can be deployed in real-world scenarios. This book also includes detailed discussions on pedestrian candidates detection, discriminative feature extraction and selection, dimension reduction, distance/metric learning, and decision/ranking enhancement.
This book is intended for professionals and researchers working in computer vision and machine learning. Advanced-level students of computer science will also find the content valuable.

The Problem of Human re-identification.- Features and Signatures.- Multi-Object Tracking.- Surveillance Camera and its Calibration.- Calibrating a Surveillance Camera Network.- Learning Viewpoint Invariant Signatures.- Learning Subject-Discriminative Features.- Dimension Reduction with Random Projections.- Sample Selection for Multi-shot Human Reidentification.- Conclusions and Future Work.
Covers every aspect of an end-to-end real-world human re-identification system
Analyzes and summarizes factors of challenges, risks and uncertainties from practical computer vision applications
Extensive evaluation and benchmarking on mainstream human re-identification algorithms and datasets   
En continuant à naviguer, vous autorisez Lavoisier à déposer des cookies à des fins de mesure d'audience. Pour en savoir plus et paramétrer les cookies, rendez-vous sur la page Confidentialité & Sécurité.