Skip to main navigation Skip to search Skip to main content

Simultaneous appearance modeling and segmentation for matching people under occlusion

  • University of Maryland, College Park

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

We describe an approach to segmenting foreground regions corresponding to a group of people into individual humans. Given background subtraction and ground plane homography, hierarchical parttemplate matching is employed to determine a reliable set of human detection hypotheses, and progressive greedy optimization is performed to estimate the best configuration of humans under a Bayesian MAP framework. Then, appearance models and segmentations are simultaneously estimated in an iterative sampling-expectation paradigm. Each human appearance is represented by a nonparametric kernel density estimator in a joint spatial-color space and a recursive probability update scheme is employed for soft segmentation at each iteration. Additionally, an automatic occlusion reasoning method is used to determine the layered occlusion status between humans. The approach is evaluated on a number of images and videos, and also applied to human appearance matching using a symmetric distance measure derived from the Kullback-Leiber divergence.

Original languageEnglish
Title of host publicationComputer Vision - ACCV 2007 - 8th Asian Conference on Computer Vision, Proceedings
PublisherSpringer Verlag
Pages404-413
Number of pages10
EditionPART 2
ISBN (Print)9783540763895
DOIs
StatePublished - 2007
Event8th Asian Conference on Computer Vision, ACCV 2007 - Tokyo, Japan
Duration: Nov 18 2007Nov 22 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume4844 LNCS

Conference

Conference8th Asian Conference on Computer Vision, ACCV 2007
Country/TerritoryJapan
CityTokyo
Period11/18/0711/22/07

Fingerprint

Dive into the research topics of 'Simultaneous appearance modeling and segmentation for matching people under occlusion'. Together they form a unique fingerprint.

Cite this