登入選單

返回Google圖書搜尋

Recognizing 3-D Objects from Range Images and Finding Their Six Pose Parameters

Jun Ohya

出版	University of Maryland, Center for Automation Research, Computer Vision Laboratory, 1991
URL	http://books.google.com.hk/books?id=DOc-HQAACAAJ&hl=&source=gbs_api

註釋

Abstract: "A method for recognizing polyhedral objects from range images and for finding their six pose parameters is studied. The approach consists of building an object library; pre-processing; hypothesis generation for the six pose parameters; and hypothesis verification. In the object library information necessary for hypothesis generation and verification is stored for each object. The pre-processing steps calculate image features at each pixel in a range image. In hypothesis generation image features are chosen at random from range images and matched to object model features. The rotation and translation required for the match are computed to yield data points in parameter spaces. The LMS (Least Median of Squares) method, a robust clustering method, generates hypotheses for object pose from the data points