Ayoub A F, Xiao Y, Khambay B, Siebert J P, Hadley D
University of Glasgow Dental School, Glasgow, UK.
Int J Oral Maxillofac Surg. 2007 May;36(5):423-8. doi: 10.1016/j.ijom.2007.02.003. Epub 2007 Apr 10.
The aim of this investigation was to assess the feasibility of building a virtual human face digitally by superimposing a photo-realistic three-dimensional (3D) soft-tissue surface on bone in the correct relationship and evaluating the registration errors associated with this method. The 3D soft-tissue surface of the face was captured using a fast stereophotogrammetry method and the underlying bone was recorded using a 3D computed tomography (CT) scanner. Using the Procrustes registration method, the outer surface of the 3D CT scan and the photo-realistic soft-tissue surfaces were merged into a single Virtual Reality Modelling Language (VRML) file and displayed using a standard VRML viewer. Quantitative measurements of registration errors were calculated in the reconstructed human head models using the signed closest point distance from the photo-realistic skin surface to the transformed CT skin surface. The registration errors between most parts of the aligned surfaces were within +/-1.5mm. The errors were relatively large around the eyebrows, eyelids and cheeks. Simultaneous recording of the face and skull may reduce this error.
本研究的目的是评估通过将逼真的三维（3D）软组织表面以正确的关系叠加在骨骼上，并评估与该方法相关的配准误差，以数字方式构建虚拟人脸的可行性。使用快速立体摄影测量法捕获面部的3D软组织表面，并使用3D计算机断层扫描（CT）扫描仪记录其下方的骨骼。使用普洛克斯配准方法，将3D CT扫描的外表面和逼真的软组织表面合并到一个虚拟现实建模语言（VRML）文件中，并使用标准的VRML查看器进行显示。在重建的人体头部模型中，使用从逼真的皮肤表面到变换后的CT皮肤表面的有符号最近点距离来计算配准误差的定量测量值。对齐表面的大部分区域之间的配准误差在+/-1.5毫米以内。眉毛、眼睑和脸颊周围的误差相对较大。同时记录面部和颅骨可能会减少这种误差。
