Effectively modeling piecewise planar urban scenes based on structure priors and CNN  

作　　者：Wei WANG Wei GAO Zhanyi HU 

机构地区：[1]School of Network Engineering, Zhoukou Normal University [2]National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences [3]University of Chinese Academy of Sciences

出　　处：《Science China(Information Sciences)》2019年第2期199-201,共3页中国科学（信息科学）（英文版）

基　　金：supported by National Key Research & Development Program of China (Grant No. 2016YFB0502002);National Natural Science Foundation of China (Grant Nos. 61333015, 61772444, 61472419);Open Project Program of the National Laboratory of Pattern Recognition (Grant No. 201700004);Natural Science Foundation of Henan Province (Grant No. 162300410347);Key Scientific and Technological Project of Henan Province (Grant No. 162102310589);College Key Research Project of Henan Province (Grant Nos. 17A520018, 17A520019)

摘　　要：Dear editor,Piecewise planar stereo methods can approximately reconstruct the complete structures of a scene by overcoming challenging difficulties (e.g.,poorly textured regions) that pixel-level stereos appear powerless. In general, these methods have three basic steps:(1) over-segmenting the image into several regions (superpixels) without overlapping;(2) generating candidate planes from initial 3D points;(3) assigning the optimal plane for each superpixel using a global method. However,,such methods can be unreliable and inefficient be-cause of three reasons:(1)inaccurate image over-segmentation(generating superpixels based only on low-level image features);(2)incomplete can-didate planes(failing to generate complete candi-date planes from sparse 3D points);(3)unreliable regularization terms(forcing two neighboring su-perpixels with similar appearances to be assigned the same plane).

关 键 词：PIECEWISE PLANAR approximately reconstruct CANDIDATE PLANES 

分 类 号：N[自然科学总论]