Real-Time GPU Surface Curvature Estimation on Deforming Meshes and Volumetric Data Sets

Abstract

Surface curvature is used in a number of areas in computer graphics, including texture synthesis and shape representation, mesh simplification, surface modeling, and nonphotorealistic line drawing. Most real-time applications must estimate curvature on a triangular mesh. This estimation has been limited to CPU algorithms, forcing object geometry to reside in main memory. However, as more computational work is done directly on the GPU, it is increasingly common for object geometry to exist only in GPU memory. Examples include vertex skinned animations and isosurfaces from GPU-based surface reconstruction algorithms. For static models, curvature can be precomputed and CPU algorithms are a reasonable choice. For deforming models where the geometry only resides on the GPU, transferring the deformed mesh back to the CPU limits performance. We introduce a GPU algorithm for estimating curvature in real time on arbitrary triangular meshes. We demonstrate our algorithm with curvature-based NPR feature lines and a curvature-based approximation for an ambient occlusion. We show curvature computation on volumetric data sets with a GPU isosurface extraction algorithm and vertex-skinned animations. We present a graphics pipeline and CUDA implementation. Our curvature estimation is up to ~18x faster than a multithreaded CPU benchmark.

Identifying surface points based on machine learning algorithms: a comprehensive analysis

“…In literature, there are several algorithms for computing curvatures of a surface (Besl and Jain, 1988; Abdelmalek, 1990; Gatzke and Grimm, 2006; Griffin et al , 2012), and most of them are based on meshed models. The principal patches, which have the curvature lines as sides of itself, were first introduced in Martin (1983).…”

Joint prediction of depths, normals and surface curvature from RGB images using CNNs

“…One technique is to simply twice-differentiate the surface [4,6], but this can lead to a high sensitivity to noise in the data and generally requires removal or rejection of surface outliers. Another technique is to estimate the surface curvature from a locally connected surface mesh based on the change in adjacent facet normal angles [7,13,14]. This method is predominantly used for computer graphics and low-noise data as it operates on a small neighbourhood of facets.…”

Parallel Marching Blocks: A Practical Isosurfacing Algorithm for Large Data on Many‐Core Architectures

“…Recent CUDA‐based MC implementations have generated indexed triangles [GWBO12, CJD15], but results have shown that their speed performance is worse than approaches that create triangle soup [NVI15b, DBG10], as generating the indexed triangles introduces severe overheads. If triangle soup is generated, a post‐processing step [Wnb12] can weld triangle vertices by eliminating redundant vertices and coincident edges, but this may introduce even more computing time and extra memory usage.…”