Foreground segmentation for static video via multi-core and multi-modal graph cut
Journal
2009 IEEE International Conference on Multimedia and Expo, ICME 2009
Pages
1362-1365
ISBN
9781424442911
Date Issued
2009
Author(s)
Chang L.-Y.
Abstract
Foreground detection is essential for semantic understanding and discovery for surveillance videos but still suffers from inefficiency and poor shape or silhouette detection. We argue to leverage multiple modalities (e.g., color appearance, foreground likelihood, spatial continuity, etc.) for foreground detection and propose a rigorous fusion method by graph cut. We further devise three strategies (e.g., dividing the graph cut problem into several subtasks, exploiting multi-core platform, etc.) to speed up the detection. Experimenting in open benchmarks, the proposed method outperforms other rival approaches in terms of detection accuracy and frame rate. ?2009 IEEE.
Subjects
Foreground detection; Graph cut; Multi-core; Silhouette; Surveillance
Type
conference paper