Paper Info
Title
Motion detection and segmentation using image mosaics
Abstract
We propose a motion segmentation algorithm for extracting foreground objects with a pan-tilt camera. Segmentation is achieved by spatio-temporal filtering of the scene to model the background. Temporal filtering is done by a set of modified AR (Auto-Regressive) filters which model the background statistics for a particular view of the scene. Backgrounds from different views of the pan-tilt camera are stitched together into a planar mosaic using a real-time image mosaicing strategy. Our algo- rithms work in real-time, require no user intervention, and facili- tate high-quality video transmission at low bandwidths.
Year
DOI
Venue
2000
10.1109/ICME.2000.871070
Multimedia and Expo, 2000. ICME 2000. 2000 IEEE International Conference
Keywords
Field
DocType
autoregressive processes,feature extraction,image segmentation,motion estimation,real-time systems,statistics,video signal processing,visual communication,autoregressive filters,background statistics,bandwidth,foreground object extraction,high-quality video transmission,image mosaics,motion detection,motion segmentation algorithm,pan-tilt camera,real-time algorithms,scene spatio-temporal filtering
Computer vision,Autoregressive model,Scale-space segmentation,Pattern recognition,Motion detection,Segmentation,Computer science,Filter (signal processing),Image segmentation,Feature extraction,Artificial intelligence,Motion estimation
Conference
Volume
ISBN
Citations 
3
0-7803-6536-4
23
PageRank 
References 
Authors
2.64
7
3
Name
Order
Citations
PageRank
Kiran S Bhat11419.40
Mahesh Saptharishi2445.49
Khosla, P.K.3931123.84