Sciweavers

Share
ICPR
2008
IEEE

Segmentation by combining parametric optical flow with a color model

9 years 9 months ago
Segmentation by combining parametric optical flow with a color model
We present a simple but efficient model for object segmentation in video scenes that integrates motion and color information in a joint probabilistic framework. Optical flow is modeled using parametric motion with Gaussian noise. The color distribution of foreground and background is described by histograms or Gaussian mixture models. Optimization is carried out using an efficient graph cut algorithm. In quantitative experiments on a variety of video data, we demonstrate that the proposed approach leads to significant reductions in error rates compared to a state-of-the-art motion-only segmentation.
Adrian Ulges, Thomas M. Breuel
Added 05 Nov 2009
Updated 06 Nov 2009
Type Conference
Year 2008
Where ICPR
Authors Adrian Ulges, Thomas M. Breuel
Comments (0)
books