<XML><RECORDS><RECORD><REFERENCE_TYPE>0</REFERENCE_TYPE><REFNUM>8034</REFNUM><AUTHORS><AUTHOR>Werghi,N.</AUTHOR><AUTHOR>Xiao,Y.</AUTHOR><AUTHOR>Siebert,J.P.</AUTHOR></AUTHORS><YEAR>2006</YEAR><TITLE>A functional-based segmentation of human body scans in arbitrary postures</TITLE><PLACE_PUBLISHED>IEEE Transactions on Systems, Man and Cybernetics: Part B, Vol. 36, No. 1</PLACE_PUBLISHED><PUBLISHER>IEEE</PUBLISHER><PAGES>153-165</PAGES><ISBN>1083-4419</ISBN><LABEL>Werghi:2006:8034</LABEL><KEYWORDS><KEYWORD>Scattered 3D range data segmentation</KEYWORD></KEYWORDS<ABSTRACT>Whole human body scanners are 3D imaging devices which have been developed to satisfy the needs of some industrial sectors (e.g. clothing industry) in terms of databases of anthropometric measurements. These scanners are capable of capturing the shape of whole body in machine readable format, thus permitting automatic extraction of the di erent body measurements. This requires the segmentation of scan data into subsets corresponding to the functional human body parts. Such a task is challenging due to the articulated and deformable nature of the human body. The attempts made so far su er from various limitations, such as being restricted to a standard speci c posture and being vulnerable to scan data artifacts. This paper presents a general framework that aims to address these challenges. A salient feature of this framework is that it can cope with various body postures and is in addition robust to noise, holes, irregular sampling and rigid transformations. Experimental results performed on real and synthetic data con rmed the validity, e ectiveness and robustness of our proposed approach.</ABSTRACT></RECORD></RECORDS></XML>