Multiview stereo and silhouette fusion via minimizing generalized reprojection error
作者：
Highlights：
• A generalized reprojection error is proposed to fuse stereo and silhouette cues.
• The method composes of convex optimization and mesh-based surface refinement.
• Insensitive to initialization and scalable for high-resolution reconstruction.
• Good performance for data with smooth texture and sparsely sampled viewpoints.
摘要
•A generalized reprojection error is proposed to fuse stereo and silhouette cues.•The method composes of convex optimization and mesh-based surface refinement.•Insensitive to initialization and scalable for high-resolution reconstruction.•Good performance for data with smooth texture and sparsely sampled viewpoints.
论文关键词：Multiview stereo,3D reconstruction,Silhouette fusion,Convex relaxation,Reprojection error
论文评审过程：Received 3 April 2014, Revised 19 October 2014, Accepted 21 October 2014, Available online 7 November 2014.
论文官网地址：https://doi.org/10.1016/j.imavis.2014.10.008
