Sarah Ali, Trung-Thien Tran, Van-Toan Cao, D. Laurendeau
{"title":"An Approach for Local Comparison of Deformable 3D Models","authors":"Sarah Ali, Trung-Thien Tran, Van-Toan Cao, D. Laurendeau","doi":"10.1109/ACPR.2013.155","DOIUrl":null,"url":null,"abstract":"In this work we present a local comparison approach of deformable 3D models. Two types of deformation are discussed. The first is volume change which is caused by either an increase or a decrease of the body volume. In the second type of deformation, the volume of the model is not altered but the change in shape results from either an articulation or a bending of parts of the object. Our strategy begins with a global alignment of the two shapes such that only their common unchanged parts will be aligned perfectly, revealing minimal distances between the parts. However, the deformed parts after alignment will be distinct and observable. In addition we provide an accurate and simple algorithm to measure the difference between the deformed parts of the models after the alignment. Experiments were conducted with real scanned 3D data in order to verify the performance of the proposed method.","PeriodicalId":365633,"journal":{"name":"2013 2nd IAPR Asian Conference on Pattern Recognition","volume":"26 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 2nd IAPR Asian Conference on Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ACPR.2013.155","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
In this work we present a local comparison approach of deformable 3D models. Two types of deformation are discussed. The first is volume change which is caused by either an increase or a decrease of the body volume. In the second type of deformation, the volume of the model is not altered but the change in shape results from either an articulation or a bending of parts of the object. Our strategy begins with a global alignment of the two shapes such that only their common unchanged parts will be aligned perfectly, revealing minimal distances between the parts. However, the deformed parts after alignment will be distinct and observable. In addition we provide an accurate and simple algorithm to measure the difference between the deformed parts of the models after the alignment. Experiments were conducted with real scanned 3D data in order to verify the performance of the proposed method.