Towards 3D human posture estimation using multiple kinects despite self-contacts

Phan, Andrew; Ferrie, Frank P.

doi:10.1109/mva.2015.7153256

Cited by 7 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The shape parameter of each subject is manually tuned before the experiments. Phan and Ferrie [13] use optical flow in the RGB domain together with depth information to perform multiview pose estimation within a human-robot collaboration context at the rate of 8 fps. They report a median joint prediction error of approximately 15 cm on a T-pose sequence.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Human Motion Capture with Multiple Depth Cameras

Shafaei

Little

2016

2016 13th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

Commonly used human motion capture systems require intrusive attachment of markers that are visually tracked with multiple cameras. In this work we present an efficient and inexpensive solution to markerless motion capture using only a few Kinect sensors. Unlike the previous work on 3d pose estimation using a single depth camera, we relax constraints on the camera location and do not assume a cooperative user. We apply recent image segmentation techniques to depth images and use curriculum learning to train our system on purely synthetic data. Our method accurately localizes body parts without requiring an explicit shape model. The body joint locations are then recovered by combining evidence from multiple views in real-time. We also introduce a dataset of~6 million synthetic depth frames for pose estimation from multiple cameras and exceed state-of-the-art results on the Berkeley MHAD dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Real-Time Human Motion Capture with Multiple Depth Cameras

Shafaei

Little

2016

2016 13th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

show abstract

“…With the shortest path algorithm, multi-touch interaction among multiple users is realized. Phan et al [ 43 ] proposed an online multi-view voting scheme (MVS) running at an interactive rate. It combined the measurement results from multiple sources to generate a fine geodesic distance graph (GDG), and then five geodesic extremes in the GDG were marked as the head, hands, and feet.…”

Section: Methods Of Point Cloud-based Joint Estimationmentioning

confidence: 99%

A Review: Point Cloud-Based 3D Human Joints Estimation

Jia

et al. 2021

Sensors

View full text Add to dashboard Cite

Joint estimation of the human body is suitable for many fields such as human–computer interaction, autonomous driving, video analysis and virtual reality. Although many depth-based researches have been classified and generalized in previous review or survey papers, the point cloud-based pose estimation of human body is still difficult due to the disorder and rotation invariance of the point cloud. In this review, we summarize the recent development on the point cloud-based pose estimation of the human body. The existing works are divided into three categories based on their working principles, including template-based method, feature-based method and machine learning-based method. Especially, the significant works are highlighted with a detailed introduction to analyze their characteristics and limitations. The widely used datasets in the field are summarized, and quantitative comparisons are provided for the representative methods. Moreover, this review helps further understand the pertinent applications in many frontier research directions. Finally, we conclude the challenges involved and problems to be solved in future researches.

show abstract

“…All the techniques mentioned so far aim at reconstructing humans by exploiting convex approximations of their shape. Phan and Ferrie [12], instead, propose a model based on geodesic distance graph model to compute a high-resolution reconstruction of humans. However, the algorithm requires a computational time of 0.6 s, that is not always suitable for applications demanding shorter reaction time, such as in HRC.…”

Section: Prior Workmentioning

confidence: 99%

Intrinsically Distributed Probabilistic Algorithm for Human–Robot Distance Computation in Collision Avoidance Strategies

et al. 2020

View full text Add to dashboard Cite

Humans and robots are becoming co-workers, both in industrial and medical applications. This new paradigm raises problems related to human safety. To accomplish and solve this issue, many researchers have developed collision avoidance strategies—mainly relying on potential field approaches—in which attractive and repulsive virtual forces are generated between manipulators and objects within a collaborative workspace. The magnitude of such virtual forces strongly depends on the relative distance between the manipulators and the approaching agents, as well on their relative velocity. In this paper, authors developed an intrinsically distributed probabilistic algorithm to compute distances between the manipulator surfaces and humans, allowing tuning the computational time versus estimation accuracy, based on the application requirements. At each iteration, the algorithm computes the human–robot distances, considering all the Cartesian points within a specific geometrical domain, built around humans’ kinematic chain, and selecting a random subset of points outside of it. Experimental validation was performed in a dynamic and unstructured condition to assess the performance of the algorithm, simulating up to six humans into the shared workspace. Tests showed that the algorithm, with the selected hardware, is able to estimate the distance between the human and the manipulator with a RMSE of 5.93 mm (maximum error of 34.86 mm).

show abstract

Towards 3D human posture estimation using multiple kinects despite self-contacts

Cited by 7 publications

References 12 publications

Real-Time Human Motion Capture with Multiple Depth Cameras

Real-Time Human Motion Capture with Multiple Depth Cameras

A Review: Point Cloud-Based 3D Human Joints Estimation

Intrinsically Distributed Probabilistic Algorithm for Human–Robot Distance Computation in Collision Avoidance Strategies

Contact Info

Product

Resources

About