Real-time three-dimensional skeletonisation using general-purpose computing on graphics processing units applied to computer vision-based human pose estimation

Abstract

Human pose estimation is the process of approximating the configuration of the body’s underlying skeletal articulation in one or more frames. The curve-skeleton of an object is a line-like representation that preserves topology and geometrical information. Finding the curve-skeleton of a volume corresponding to the person is a good starting point for approximating the underlying skeletal structure. In this paper, a GPU implementation of a fully parallel thinning algorithm based on the critical kernel framework is presented. The algorithm is compared to three other state-of-the-art skeletonisation methods—two CPU and one GPU implementation—using both real and synthetic data. It is demonstrated that all four achieve close to real-time frame rates, however, the proposed algorithm yields superior accuracy and robustness when used in a pose estimation context. The GPU implementation is > 8× faster than a CPU implementation of the same algorithm, and the positions of the 4 extremities are estimated with rms error ∼6 cm and ∼98% of frames correctly labelled for some sequences.

A New Method of Central Axis Extracting for Pore Network Modeling in Rock Engineering

“…Extraction of Pore Axis. The theory and criterion proposed by Bakken and Eliassen [45], Kong and Rosenfeld [46], and Ma [47] lay the foundation for the pore axis extraction. Later, Palágyi and Kuba [34] developed a refinement algorithm based on delete template.…”

“…When voxels of one phase have the maximum number of connected voxels, i.e., 26, they may connect with each other in the form of a point or a line, which do not occur between solid-phase voxels. Considering rock skeleton is solid and the simple point described by Bakken and Eliassen [45], in this work, we define that the pore voxel is connected with 26 nearby pore voxels at most and the rock skeleton voxel is connected with 6 nearby skeleton voxels at most. Therefore, the condition to delete black point depends on the relationship in its 26-neighborhood voxels.…”