Limits...
Articulated Non-Rigid Point Set Registration for Human Pose Estimation from 3D Sensors.

Ge S, Fan G - Sensors (Basel) (2015)

Bottom Line: We introduce a visible point extraction method to initialize a new template for the current frame from the previous frame, which effectively reduces the ambiguity and uncertainty during registration.Third, to support robust and stable pose tracking, we develop a segment volume validation technique to detect tracking failures and to re-initialize pose registration if needed.The experimental results on both benchmark 3D laser scan and depth datasets demonstrate the effectiveness of the proposed framework when compared with state-of-the-art algorithms.

View Article: PubMed Central - PubMed

Affiliation: School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74078, USA. song.ge@okstate.edu.

ABSTRACT
We propose a generative framework for 3D human pose estimation that is able to operate on both individual point sets and sequential depth data. We formulate human pose estimation as a point set registration problem, where we propose three new approaches to address several major technical challenges in this research. First, we integrate two registration techniques that have a complementary nature to cope with non-rigid and articulated deformations of the human body under a variety of poses. This unique combination allows us to handle point sets of complex body motion and large pose variation without any initial conditions, as required by most existing approaches. Second, we introduce an efficient pose tracking strategy to deal with sequential depth data, where the major challenge is the incomplete data due to self-occlusions and view changes. We introduce a visible point extraction method to initialize a new template for the current frame from the previous frame, which effectively reduces the ambiguity and uncertainty during registration. Third, to support robust and stable pose tracking, we develop a segment volume validation technique to detect tracking failures and to re-initialize pose registration if needed. The experimental results on both benchmark 3D laser scan and depth datasets demonstrate the effectiveness of the proposed framework when compared with state-of-the-art algorithms.

No MeSH data available.


Ground-truth generation of joint positions for SCAPE data.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541828&req=5

f8-sensors-15-15218: Ground-truth generation of joint positions for SCAPE data.

Mentions: The SCAPE dataset contains a series of 3D scan data captured from one male subject (the only one publicly available under different poses), which are fully registered (the index of each point stays the same across all poses). It has one initial pose with ground-truth joint positions. To perform quantitative comparative analysis, we develop a simple, yet effective four-step approach to generate the ground-truth joint positions for all other poses, as shown in Figure 8. First, we perform body segmentation for the initial pose according to joint positions. Second, for each joint, we find a set of neighboring points around the joint area between two connected body segments and compute LLE weight coefficients to represent each joint locally. Third, we transfer the segmental labels from the standard pose for any new target pose. Fourth, we use LLE weight coefficients and the associated neighboring points, which share the same indexes as those in the initial pose, to reconstruct each joint position in the target pose. In this way, all poses will have the ground-truth joint positions created for performance evaluation.


Articulated Non-Rigid Point Set Registration for Human Pose Estimation from 3D Sensors.

Ge S, Fan G - Sensors (Basel) (2015)

Ground-truth generation of joint positions for SCAPE data.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541828&req=5

f8-sensors-15-15218: Ground-truth generation of joint positions for SCAPE data.
Mentions: The SCAPE dataset contains a series of 3D scan data captured from one male subject (the only one publicly available under different poses), which are fully registered (the index of each point stays the same across all poses). It has one initial pose with ground-truth joint positions. To perform quantitative comparative analysis, we develop a simple, yet effective four-step approach to generate the ground-truth joint positions for all other poses, as shown in Figure 8. First, we perform body segmentation for the initial pose according to joint positions. Second, for each joint, we find a set of neighboring points around the joint area between two connected body segments and compute LLE weight coefficients to represent each joint locally. Third, we transfer the segmental labels from the standard pose for any new target pose. Fourth, we use LLE weight coefficients and the associated neighboring points, which share the same indexes as those in the initial pose, to reconstruct each joint position in the target pose. In this way, all poses will have the ground-truth joint positions created for performance evaluation.

Bottom Line: We introduce a visible point extraction method to initialize a new template for the current frame from the previous frame, which effectively reduces the ambiguity and uncertainty during registration.Third, to support robust and stable pose tracking, we develop a segment volume validation technique to detect tracking failures and to re-initialize pose registration if needed.The experimental results on both benchmark 3D laser scan and depth datasets demonstrate the effectiveness of the proposed framework when compared with state-of-the-art algorithms.

View Article: PubMed Central - PubMed

Affiliation: School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74078, USA. song.ge@okstate.edu.

ABSTRACT
We propose a generative framework for 3D human pose estimation that is able to operate on both individual point sets and sequential depth data. We formulate human pose estimation as a point set registration problem, where we propose three new approaches to address several major technical challenges in this research. First, we integrate two registration techniques that have a complementary nature to cope with non-rigid and articulated deformations of the human body under a variety of poses. This unique combination allows us to handle point sets of complex body motion and large pose variation without any initial conditions, as required by most existing approaches. Second, we introduce an efficient pose tracking strategy to deal with sequential depth data, where the major challenge is the incomplete data due to self-occlusions and view changes. We introduce a visible point extraction method to initialize a new template for the current frame from the previous frame, which effectively reduces the ambiguity and uncertainty during registration. Third, to support robust and stable pose tracking, we develop a segment volume validation technique to detect tracking failures and to re-initialize pose registration if needed. The experimental results on both benchmark 3D laser scan and depth datasets demonstrate the effectiveness of the proposed framework when compared with state-of-the-art algorithms.

No MeSH data available.