抄録
In this paper, a novel approach is proposed to recover human body pose from 3D voxel data. The use of voxel data leads to viewpoint-free estimation, which benefits in that reconstruction of a training model is needless in different multi-camera arrangements. The chief advantage of our approach is speed, which enables real-time processing when capturing 8 VGA size images in 30 [fps] . Our approach is mainly based on an example-based approach. Human posture candidates are constructed beforehand, and the most appropriate posture candidate is selected per frame by comparing the similarity between 3D voxel data and posture candidates. Derivation of similarity is formulated by introducing a histogram-based feature vector that represents the 3D context of human body. In addition, a fast near-neighbor search metric is installed prior to the evaluation process, to reduce the computational cost and ensure real-time processing. Estimation stability is also improved by a motion graph, which adds a smoothing effect to the motion sequence. We demonstrate the effectiveness of our approach with experiments on both synthetic and real image sequences.
