Articulated and Generalized Gaussian Kernel Correlation for Human Pose Estimation

• Published in: IEEE transactions on image processing Vol. 25; no. 2; pp. 776 - 789
• Main Authors: Ding, Meng, Fan, Guoliang
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.02.2016; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Articulated Pose Estimation; Computational modeling; Correlation; Depth Sensor; Gaussian; Hand - physiology; Hand Pose Tracking; Human Pose Tracking; Humans; Image Processing, Computer-Assisted - methods; Kernel; Kernel Correlation; Kernels; Kinect; Mathematical models; Normal Distribution; Posture - physiology; Robustness; Shape; Shape Modeling; Similarity; Skeleton; Sum of Gaussians (SoG); Tracking; Visibility; sum of Gaussians (SoG); articulated pose estimation; hand pose tracking; shape modeling; depth sensor; human pose tracking; kinect; Kernel correlation
• ISSN: 1057-7149; 1941-0042
• DOI: 10.1109/TIP.2015.2507445

In this paper, we propose an articulated and generalized Gaussian kernel correlation (GKC)-based framework for human pose estimation. We first derive a unified GKC representation that generalizes the previous sum of Gaussians (SoG)-based methods for the similarity measure between a template and an observation both of which are represented by various SoG variants. Then, we develop an articulated GKC (AGKC) by integrating a kinematic skeleton in a multivariate SoG template that supports subject-specific shape modeling and articulated pose estimation for both the full body and the hands. We further propose a sequential (body/hand) pose tracking algorithm by incorporating three regularization terms in the AGKC function, including visibility, intersection penalty, and pose continuity. Our tracking algorithm is simple yet effective and computationally efficient. We evaluate our algorithm on two benchmark depth data sets. The experimental results are promising and competitive when compared with the state-of-the-art algorithms.