Abstract | ||
---|---|---|
We address the elusive goal of estimating optical flow both accurately and efficiently by adopting a sparse-to-dense approach. Given a set of sparse matches, we regress to dense optical flow using a learned set of full-frame basis flow fields. We learn the principal components of natural flow fields using flow computed from four Hollywood movies. Optical flow fields are then compactly approximated as a weighted sum of the basis flow fields. Our new PCA-Flow algorithm robustly estimates these weights from sparse feature matches. The method runs in under 200ms/frame on the MPI-Sintel dataset using a single CPU and is more accurate and significantly faster than popular methods such as LDOF and Classic+NL. For some applications, however, the results are too smooth. Consequently, we develop a novel sparse layered flow method in which each layer is represented by PCA-Flow. Unlike existing layered methods, estimation is fast because it uses only sparse matches. We combine information from different layers into a dense flow field using an image-aware MRF. The resulting PCA-Layers method runs in 3.2s/frame, is significantly more accurate than PCA-Flow, and achieves state-of-the-art performance in occluded regions on MPI-Sintel. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/CVPR.2015.7298607 | 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) |
Keywords | Field | DocType |
sparse-to-dense optical flow field estimation,full-frame basis flow field learning,principal component learning,natural flow fields,Hollywood movies,weighted sum,PCA-Flow algorithm,robust weight estimation,sparse feature matches,MPI-Sintel dataset,CPU,sparse-layered flow method,layer representation,image-aware MRF,PCA-Layers method | Computer vision,Pattern recognition,Computer science,Sparse approximation,Flow (psychology),Optical flow estimation,Artificial intelligence,Optical flow,Principal component analysis | Conference |
Volume | Issue | ISSN |
2015 | 1 | 1063-6919 |
Citations | PageRank | References |
28 | 0.91 | 47 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jonas Wulff | 1 | 438 | 17.59 |
Michael J. Black | 2 | 11233 | 1536.41 |