A Survey on Deep Geometry Learning: From a Representation Perspective

Abstract

Researchers have now achieved great success on dealing with 2D images using deep learning. In recent years, 3D computer vision and Geometry Deep Learning gain more and more attention. Many advanced techniques for 3D shapes have been proposed for different applications. Unlike 2D images, which can be uniformly represented by regular grids of pixels, 3D shapes have various representations, such as depth and multi-view images, voxel-based representation, point-based representation, mesh-based representation, implicit surface representation, etc. However, the performance for different applications largely depends on the representation used, and there is no unique representation that works well for all applications.Therefore, in this survey, we review recent development in deep learning for 3D geometry from a representation perspective, summarizing the advantages and disadvantages of different representations in different applications. We also present existing datasets in these representations and further discuss future research directions.

Deep learning frameworks for point cloud reconstruction

“…They map a 3D point to a real value instead of a probability, indicating the spatial relation and distance to the 3D surface. If SDF (x) is the signed distance value of a given 3D point x ∈ R 3 , then SDF (x) > 0 if point x is outside the 3D shape, SDF (x) < 0 if point x is inside the shape, and SDF (x) = 0 if point x is on the surface [85]. The absolute value of SDF (x) gives the distance between point x and the surface.…”

“…Earlier datasets like the NYU Depth dataset [51] used RGB-D images collected from depth sensors such as the Microsoft Kinect. ScanNet was another dataset that extended RGB-D to videos with 3D camera poses of approximately 1500 scenes for surface reconstruction and semantic segmentation [85].…”