000199737 001__ 199737
000199737 005__ 20190316235926.0
000199737 037__ $$aCONF
000199737 245__ $$aREMODE: Probabilistic, Monocular Dense Reconstruction in Real Time
000199737 269__ $$a2014
000199737 260__ $$c2014
000199737 336__ $$aConference Papers
000199737 520__ $$aIn this paper, we solve the problem of estimating dense and accurate depth maps from a single moving camera. A probabilistic depth measurement is carried out in real time on a per-pixel basis and the computed uncertainty is used to reject erroneous estimations and provide live feedback on the reconstruction progress. Our contribution is a novel approach to depth map computation that combines Bayesian estimation and recent development on convex optimization for image processing. We demonstrate that our method outperforms state-of-the-art techniques in terms of accuracy, while exhibiting high efficiency in memory usage and computing power. We call our approach REMODE (REgularized MOnocular Depth Estimation) and the CUDA-based implementation runs at 30Hz on a laptop computer.
000199737 700__ $$aPizzoli, Matia
000199737 700__ $$aForster, Christian
000199737 700__ $$aScaramuzza, Davide
000199737 7112_ $$dMay 31 - June 7, 2014$$cHong Kong, China$$a2014 IEEE International Conference on Robotics and Automation (ICRA 2014)
000199737 8564_ $$uhttps://infoscience.epfl.ch/record/199737/files/ICRA14_Pizzoli.pdf$$zn/a$$s3333158$$yn/a
000199737 909C0 $$xU12367$$0252409$$pNCCR-ROBOTICS
000199737 909CO $$qGLOBAL_SET$$pconf$$ooai:infoscience.tind.io:199737
000199737 917Z8 $$x221818
000199737 937__ $$aEPFL-CONF-199737
000199737 973__ $$rREVIEWED$$sPUBLISHED$$aOTHER
000199737 980__ $$aCONF