Share Email Print

Proceedings Paper

A trajectory based video segmentation for surveillance applications
Author(s): Naveen M. Thomas; Nishan Canagarajah
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Video segmentation for content based retrieval has traditionally been done using shot cut detection algorithms that search for abrupt changes in scene content. Surveillance videos however, usually use still cameras, and do not contain any shots. Hence, a novel high level semantic change detection algorithm is proposed in this paper that uses object trajectory features to segment surveillance footage. These trajectory features are extracted automatically, using background subtraction and a multiple blob tracking algorithm. The trajectory features are first used to remove false object detections from background subtraction. Semantics extracted from the remaining object trajectories are then used to segment the video. The results of the algorithm when applied to surveillance data are compared with hand labeled segmentation to obtain precision recall curves and harmonic mean. Comparisons with traditional background subtraction and video segmentation algorithms show a drastic improvement in performance.

Paper Details

Date Published: 29 January 2007
PDF: 10 pages
Proc. SPIE 6506, Multimedia Content Access: Algorithms and Systems, 65060A (29 January 2007); doi: 10.1117/12.703840
Show Author Affiliations
Naveen M. Thomas, Univ. of Bristol (United Kingdom)
Nishan Canagarajah, Univ. of Bristol (United Kingdom)

Published in SPIE Proceedings Vol. 6506:
Multimedia Content Access: Algorithms and Systems
Alan Hanjalic; Raimondo Schettini; Nicu Sebe, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?