Model Refinement by Iterative Normal-From-Shading
暂无分享,去创建一个
The surface normals of a 3D human model, generated form multiple viewpoint capture, are refined using an iterative variant of the shape-from-shading technique to recover fine details of clothing enabling relighting of the model according to a new virtual environment. The method requires the images to consist of a number of uniformly coloured surface patches.
[1] John Hart,et al. Textureshop: texture synthesis as a photograph editing tool , 2004, SIGGRAPH 2004.
[2] Adrian Hilton,et al. Volumetric Stereo with Silhouette and Feature Constraints , 2006, BMVC.