2017/10/27 by René Schuster, Oliver Wasenmüller, Schuster, René +7
Computer Science · #Advanced Image Processing Techniques #Advanced Vision and Imaging #Computer Graphics and Visualization Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences
paper · pdf · doi:10.48550/arxiv.1710.10096
openalex publication_date 2017/10/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
While most scene flow methods use either variational optimization or a strong\nrigid motion assumption, we show for the first time that scene flow can also be\nestimated by dense interpolation of sparse matches. To this end, we find sparse\nmatches across two stereo image pairs that are detected without any prior\nregularization and perform dense interpolation preserving geometric and motion\nboundaries by using edge information. A few iterations of variational energy\nminimization are performed to refine our results, which are thoroughly\nevaluated on the KITTI benchmark and additionally compared to state-of-the-art\non MPI Sintel. For application in an automotive context, we further show that\nan optional ego-motion model helps to boost performance and blends smoothly\ninto our approach to produce a segmentation of the scene into static and\ndynamic parts.\n