Laplacian meshes for monocular 3d shape recovery

Jonas Östlund,Aydin Varol,Dat Tien Ngo,Pascal Fua

COMPUTER VISION - ECCV 2012, PT III（2012）

引用 67|浏览0

暂无评分

摘要

We show that by extending the Laplacian formalism, which was first introduced in the Graphics community to regularize 3D meshes, we can turn the monocular 3D shape reconstruction of a deformable surface given correspondences with a reference image into a well-posed problem. Furthermore, this does not require any training data and eliminates the need to pre-align the reference shape with the one to be reconstructed, as was done in earlier methods.

查看译文

关键词

training data,well-posed problem,reference shape,laplacian mesh,deformable surface,shape reconstruction,graphics community,reference image,earlier method,shape recovery,laplacian formalism,computer vision

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要