Font Size: a A A

Research And Implementation Of Facial Reconstruction Based On A Regional Statistical Craniofacial Model

Posted on:2011-05-11Degree:MasterType:Thesis
Country:ChinaCandidate:Y F ZhangFull Text:PDF
GTID:2178360305959639Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Reconstruction of facial soft tissue is an essential processing phase in forensics, anthropology as well as maxillofacial surgery. In this paper, we propose a face appearance reconstruction algorithm based on a Regional Statistical Craniofacial model called RSCM, which can cut down the reconstruction error caused by the small sample size problem. After the shape of the craniofacial model is decomposed into a few of segments, such as the eyes, the nose and the mouth regions, the joint statistical models of different regions are constructed independently. Then the face reconstruction task is formulated as a miss data problem, and is also fulfilled region by region respectively. Finally, the recovered face is assembled by less error regions. The main contents of this paper are listed as follow:(1) 3D craniofacial preprocessing:3D craniofacial models is acquired through a developed snake algorithm, then ICP registration is applied on each one, finally correspondence is established by minimum distance method.(2) Proposed A Decomposition and Assemble algorithm on craniofacial:The decomposition is executed through an assistant geometric template defined by the landmarks on face and skull model. On the contrary, assembling method is proposed for solving the not smooth assembled face may appear on regions edge, made the result face more nature.(3) Regional Statistical Craniofacial Model (RSCM):Using regional face and skull data, the statistical model is constructed respectively for each region. Since the number of vertices in each region is smaller than the whole model, RSCM is more flexible than Integrated Statistical Craniofacial Model.(4) Regional Statistical Craniofacial Model based facial reconstruction:The regions of recovered faces are acquired respectively by solving a formulated missing data problem. Then assembling algorithm is applied on those to achieve a completed smooth face model. And the result is proved better than integrated statistical model.(5) Facial Reconstruction System Development:A Statistical based Craniofacial Reconstruction system is designed and implement to cater the demand of practice.
Keywords/Search Tags:Craniofacial reconstruction, 3D model decomposition, Regional Statistical Model, Regions assembling
PDF Full Text Request
Related items