LACHAUD , Jacques-Olivier

In many geometry processing applications, differential geometric quantities estimation such as curvature or normal vector field is an essential step. In [1], we have defined curvature estimators on digital shape boundaries based on Integral Invariants. In this paper, we focus on implementation details of these estimators.

Recent works on data-driven sketch-based modeling use either voxel grids or normal/depth maps as geometric representations compatible with convolutional neural networks. While voxel grids can represent complete objects-including parts not visible in the sketches-their memory consumption restricts them to low-resolution predictions. In contrast,...