 |
Technical Program
Paper Detail
| Paper: | WP-P8.5 |
| Session: | Image/Video Processing Applications: Models and Methods |
| Time: | Wednesday, October 11, 14:20 - 17:00 |
| Presentation: |
Poster
|
| Title: |
REGION-BASED STATISTICAL BACKGROUND MODELING FOR FOREGROUND OBJECT SEGMENTATION |
| Authors: |
Kristof Op De Beeck; Katholieke Universiteit Leuven | | | | | Irene Y.H. Gu; Chalmers University of Technology | | | | | Liyuan Li; Institute for Infocomm Research | | | | | Mats Viberg; Chalmers University of Technology | | | | | Bart De Moor; Katholieke Universiteit Leuven | | |
| Abstract: |
This paper proposes a novel region-based scheme for dynamically modeling time-evolving statistics of video background, leading to an effective segmentation of foreground moving objects for a video surveillance system. In [1] statistical-based video surveillance systems employ a Bayes decision rule for classifying foreground and background changes in individual pixels. Although principal feature representations significantly reduce the size of tables of statistics, pixel-wise maintenance remains a challenge due to the computations and memory requirement. The proposed region-based scheme, which is extension of the above method, replaces pixel-based statistics by region-based statistics through introducing dynamic background region (or pixel) merging and splitting. Simulations have been conducted to several outdoor and indoor image sequences, and results have shown a significant reduction of memory requirements for tables of statistics while maintaining relatively good quality in foreground segmented video objects. |
|
