000128498 001__ 128498
000128498 005__ 20181114182058.0
000128498 037__ $$aCONF
000128498 245__ $$aWide-baseline Stereo from Multiple Views: a Probabilistic Account
000128498 269__ $$a2004
000128498 260__ $$c2004
000128498 336__ $$aConference Papers
000128498 520__ $$aThis paper describes a method for dense depth reconstruction from a small set of wide-baseline images. In a widebaseline setting an inherent difficulty which complicates the stereo-correspondence problem is self-occlusion. Also, we have to consider the possibility that image pixels in different images, which are projections of the same point in the scene, will have different color values due to non-Lambertian effects or discretization errors. We propose a Bayesian approach to tackle these problems. In this framework, the images are regarded as noisy measurements of an underlying ’true’ image-function. Also, the image data is considered incomplete, in the sense that we do not know which pixels from a particular image are occluded in the other images. We describe an EM-algorithm, which iterates between estimating values for all hidden quantities, and optimizing the current depth estimates. The algorithm has few free parameters, displays a stable convergence behavior and generates accurate depth estimates. The approach is illustrated with several challenging real-world examples. We also show how the algorithm can generate realistic view interpolations and how it merges the information of all images into a new, synthetic view.
000128498 6531_ $$amulti-view stereo
000128498 6531_ $$anovel view generation
000128498 700__ $$0244088$$aStrecha, Christoph$$g182325
000128498 700__ $$aFransen, Rik
000128498 700__ $$aVan Gool, Luc
000128498 7112_ $$aIEEE Conference on Computer Vision and Pattern Recognition$$cWashington$$dJune 27 - July 2, 2004 
000128498 773__ $$j2$$q552-559$$tProceedings of the IEEE Conference on Computer Vision and Pattern Recognition
000128498 8564_ $$uhttp://cvl.umiacs.umd.edu/conferences/cvpr2004/$$zURL
000128498 8564_ $$s992563$$uhttps://infoscience.epfl.ch/record/128498/files/cvpr_2004.pdf$$zn/a
000128498 909C0 $$0252087$$pCVLAB$$xU10659
000128498 909CO $$ooai:infoscience.tind.io:128498$$pconf$$pIC$$pGLOBAL_SET
000128498 937__ $$aCVLAB-CONF-2008-024
000128498 973__ $$aOTHER$$rREVIEWED$$sPUBLISHED
000128498 980__ $$aCONF