83
Voted
IJCV
2008
14 years 8 months ago
2008
We present an unsupervised approach for learning a generative layered representation of a scene from a video for motion segmentation. The learnt model is a composition of layers, ...