融合部分深度线索的立体匹配方法.pdfVIP

() , 45 , 5 ,2009 9 Acta Scientiarum Naturalium Universitatis Pekinensis, Vol.45, No. 5 (Sept. 2009) 融合部分深度线索的立体匹配方法 1, 2,­ 2 马祥音 查红彬 11 , 310023; 21 , 100871; ­ E-mail: maxiangyin1983@ gmail. com , , ,, , , , ; ; ; TP391 Stereo Matching by Incorporating Depth Cues 1,2, ­ 2 MA Xiangyin , ZHA Hongbin 11 College of Computer Science andTechnology, Zhejiang University of Technology, Hangzhou 310023; 21 Key Laboratory on Machine Perception (Ministry of Education) , Peking University, eijing 100871; ­ E-mail: maxiangyin1983@ gmail. com Abstract Focusing on scenes of outdoor buildings, the authors present graph-based stereo matching methods that incorporate depth cues to acquire more accurate disparity maps. Firstly, given a portion of scan data, planar disparity layers, which correspond to 3D planes, are precisely extracted to compose the label set. After that, the graph-basedmatching algorithm can be formulated inthe color segment domain insteadof pixel domain. Furthermore, in order to avoid the inconvenient scanning for every scene, the scanner is only used to create a training set consisting of image-depthmap pairs. Then the depth cues, predicted through training a Markov Random Field to model the relationship between the depth and the image features, can still serve as extra constraint for the correspondence problem. The experimental results show the convincing performance of proposed methods in the reconstruction of different scenes. Key words stereo matching; planar disparity layer; Markov Random Field; statistical learning , , ,