Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow

CVPR 2014

doi:10.1109/CVPR.2014.452 /cvpr/2014/bao2014cvpr-fast/

Abstract

We present a fast optical flow algorithm that can handle large displacement motions. Our algorithm is inspired by recent successes of local methods in visual correspondence searching as well as approximate nearest neighbor field algorithms. The main novelty is a fast randomized edge-preserving approximate nearest neighbor field algorithm which propagates self-similarity patterns in addition to offsets. Experimental results on public optical flow benchmarks show that our method is significantly faster than state-of-the-art methods without compromising on quality, especially when scenes contain large motions.

PDF CVPR Semantic Scholar

Cite

Text

Bao et al. "Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow." Conference on Computer Vision and Pattern Recognition, 2014. doi:10.1109/CVPR.2014.452

Markdown

[Bao et al. "Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow." Conference on Computer Vision and Pattern Recognition, 2014.](https://mlanthology.org/cvpr/2014/bao2014cvpr-fast/) doi:10.1109/CVPR.2014.452

BibTeX

@inproceedings{bao2014cvpr-fast,
  title     = {{Fast Edge-Preserving PatchMatch for Large Displacement Optical Flow}},
  author    = {Bao, Linchao and Yang, Qingxiong and Jin, Hailin},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2014},
  doi       = {10.1109/CVPR.2014.452},
  url       = {https://mlanthology.org/cvpr/2014/bao2014cvpr-fast/}
}