Abstract

We present an algorithm for extracting high quality tem- porally coherent alpha mattes of ob jects from a video. Our approach extends the conventional image matting approach, i.e. closed-form mat- ting, to video by using multi-frame nonlocal matting Laplacian. Our multi-frame nonlocal matting Laplacian is defined over a nonlocal neigh- borhood in spatial temporal domain, and it solves the alpha mattes of several video frames all together simultaneously. To speed up computa- tion and to reduce memory requirement for solving the multi-frame non- local matting Laplacian, we use the approximate nearest neighbor(ANN) to find the nonlocal neighborhood and the k-d tree implementation to di- vide the nonlocal matting Laplacian into several smaller linear systems. Finally, we adopt the nonlocal mean regularization to enhance tempo- ral coherence of the estimated alpha mattes and to correct alpha matte errors at low contrast regions. We demonstrate the effectiveness of our approach on various examples with qualitative comparisons to the results from previous matting algorithms.