Share Email Print

Proceedings Paper

EM motion segmentation based on MRF model
Author(s): Jie Wei; Izidor Gertner
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In our earlier work, a Two-Pass motion estimation Algorithm (TPA) was developed to estimate a motion field for two adjacent frames in an image sequence where contextural constraints are handeled by several Markov Random Fields (MRFs) and the most A Posteriori (MAP) configuration is taken to be the resulting motion field. Currently in the disciplines of digital library and video processing of utmost interest are the extraction and representation of visual objects. Instead of estimating motion field, in this paper we focus on segmenting out visual objects based on spatial and temporal properties present in two contiguous frames under the MRF-MAP-MFT scheme. To achieve object segmentation, within the framework of EM optimization a novel concept "motion boundary field" is introduced which can turn off interactions between different object regions and in the mean time remove spurious objerct boundaries. Furthermore, in light of the generally smooth and slow velocities in-between two contiguous frames, we found that in the process of calculating matching blocks, assigning different weights to different locations can result in better object segmentation.

Paper Details

Date Published: 16 September 2003
PDF: 9 pages
Proc. SPIE 5094, Automatic Target Recognition XIII, (16 September 2003); doi: 10.1117/12.499593
Show Author Affiliations
Jie Wei, CUNY, City College (United States)
Izidor Gertner, CUNY, City College (United States)

Published in SPIE Proceedings Vol. 5094:
Automatic Target Recognition XIII
Firooz A. Sadjadi, Editor(s)

© SPIE. Terms of Use
Back to Top