RT Conference Proceedings
T1 Motion Cooperation: Smooth Piece-Wise Rigid Scene Flow from RGB-D Images
A1 Jaimez, Mariano
A1 Souiai, Mohamed
A1 Stuckler, Jorg
A1 González-Jiménez, Antonio Javier
A1 Cremers, Daniel
K1 Visión por ordenador
AB We propose a novel joint registration and segmentation approach to estimate scene flow from RGB-D images. Instead of assuming the scene to be composed of a number of independent rigidly-moving parts, we use non-binary labels to capture non-rigid deformations at transitions betweenthe rigid parts of the scene. Thus, the velocity of any point can be computed as a linear combination (interpolation) of the estimated rigid motions, which provides better resultsthan traditional sharp piecewise segmentations. Within a variational framework, the smooth segments of the scene and their corresponding rigid velocities are alternately refineduntil convergence. A K-means-based segmentation is employed as an initialization, and the number of regions is subsequently adapted during the optimization process to capture any arbitrary number of independently moving objects.We evaluate our approach with both synthetic andreal RGB-D images that contain varied and large motions. The experiments show that our method estimates the scene flow more accurately than the most recent works in the field, and at the same time provides a meaningful segmentation of the scene based on 3D motion.
YR 2016
FD 2016-01-07
LK http://hdl.handle.net/10630/10863
UL http://hdl.handle.net/10630/10863
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.         Spanish Government under the grant programs FPI-MICINN 2012 and DPI2014- 55826-R (co-founded by the European Regional Development Fund), as well as by the EU ERC grant Convex Vision (grant agreement no. 240168).
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 24 ene 2026