The dense estimation of motion and appearance in layers

Hulya Yalcin, Michael J. Black, Ronan Fablet

Research output: Contribution to journalConference articlepeer-review

12 Citations (Scopus)

Abstract

Segmenting image sequences into meaningful layers is fundamental to many applications such as surveillance, tracking, and video summarization. Background subtraction techniques are popular for their simplicity and, while they provide a dense (pixelwise) estimate of foreground/background, they typically ignore image motion which can provide a rich source of information about scene structure. Conversely, layered motion estimation techniques typically ignore the temporal persistence of image appearance and provide parametric (rather than dense) estimates of optical flow. Recent work adaptively combines motion and appearance estimation in a mixture model framework to achieve robust tracking. Here we extend mixture model approaches to cope with dense motion and appearance estimation. We develop a unified Bayesian framework to simultaneously estimate the appearance of multiple image layers and their corresponding dense flow fields from image sequences. Both the motion and appearance models adapt over time and the probabilistic formulation can be used to provideasegmentation of thescene into foreground/background regions. This extension of mixture models includes prior probability models for the spatial and temporal coherence of motion and appearance. Experimental results show that the simultaneous estimation of appearance models and flow fields in multiple layers improves the estimation of optical flow at motion boundaries.

Original languageEnglish
Article number1384964
JournalIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
Volume2004-January
Issue numberJanuary
DOIs
Publication statusPublished - 2004
Externally publishedYes
Event2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2004 - Washington, United States
Duration: 27 Jun 20042 Jul 2004

Bibliographical note

Publisher Copyright:
© 2004 IEEE.

Fingerprint

Dive into the research topics of 'The dense estimation of motion and appearance in layers'. Together they form a unique fingerprint.

Cite this