Abstract
Person re-identification is defined as the problem of recognizing an individual captured in diverse times and/or locations over several nonoverlapping camera views, considering a large set of candidates. This problem affects primarily the management of distributed, multiview surveillance systems, in which subjects must be tracked across different places, either a posteriori or on-the-fly when they move through different locations. Re-identification is a very difficult problem, as most of the time people can be captured by several low resolution cameras, under occlusion conditions, badly (and different from view to view) illuminated, and in varying poses. In this context, a robust modeling of the entire body appearance of a person is necessary, especially when other classical biometric cues (face, gait) are not available or difficult to catch, due to the sensors’ scarce resolution or low frame-rate. This chapter gives an overview of the re-identification problem, illustrating the standard re-identification pipeline and detailing the several approaches and techniques, devoting more attention to those which have shown to be particularly effective and significant.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Academic Press Library in Signal Processing, Volume 6
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
