Sciweavers

CVPR
2004
IEEE

Video Repairing: Inference of Foreground and Background under Severe Occlusion

15 years 26 days ago
Video Repairing: Inference of Foreground and Background under Severe Occlusion
In this paper, we propose a new method, video repairing, to robustly infer missing static background and moving foreground due to severe damage or occlusion from a video. To recover background pixels, we extend the image repairing method, where layer segmentation and homography blending are used to preserve temporal coherence and avoid flickering. By exploiting the constraint imposed by periodic motion and a subclass of camera and object motions, we adopt a two-phase approach to repair moving foreground pixels: In the sampling phase, motion data are sampled and regularized by 3D tensor voting to maintain temporal coherence and motion periodicity. In the alignment phase, missing moving foreground pixels are inferred by spatial and temporal alignment of the sampled motion data at multiple scales. We experimented our system with some difficult examples, where the camera can be stationary or in motion.
Jiaya Jia, Tai-Pang Wu, Yu-Wing Tai, Chi-Keung Tan
Added 12 Oct 2009
Updated 29 Oct 2009
Type Conference
Year 2004
Where CVPR
Authors Jiaya Jia, Tai-Pang Wu, Yu-Wing Tai, Chi-Keung Tang
Comments (0)