2 Hits in 2.3 sec

Highly Overparameterized Optical Flow Using PatchMatch Belief Propagation [chapter]

Michael Hornáček, Frederic Besse, Jan Kautz, Andrew Fitzgibbon, Carsten Rother
2014 Lecture Notes in Computer Science  
We show that optimization over this high-dimensional, continuous state space can be carried out using an adaptation of the recently introduced PatchMatch Belief Propagation (PMBP) energy minimization algorithm  ...  We propose to compute optical flow using what is ostensibly an extreme overparameterization: depth, surface normal, and frame-to-frame 3D rigid body motion at every pixel, giving a total of 9 DoF.  ...  For optical flow from a pair of images without strictly enforcing epipolar geometry, we show that PatchMatch Belief Propagation (PMBP) of Besse et al.  ... 
doi:10.1007/978-3-319-10578-9_15 fatcat:hfu7h2wnejdunhmujraupgr5mi

Patch Match Filter: Efficient Edge-Aware Filtering Meets Randomized Search for Fast Correspondence Field Estimation

Jiangbo Lu, Hongsheng Yang, Dongbo Min, Minh N. Do
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
Focusing on dense correspondence field estimation in this paper, we demonstrate PMF's applications in stereo and optical flow.  ...  When the label space is huge, which is often the case for (subpixelaccurate) stereo and optical flow estimation, their computational complexity becomes quickly unacceptable.  ...  Optical Flow Results We evaluate our PMF-OF methods using the Middlebury flow benchmark [1].  ... 
doi:10.1109/cvpr.2013.242 dblp:conf/cvpr/LuYMD13 fatcat:al6bsahj2redfha76on5pcdngy