MARKERLESS HUMAN MOTION TRACKING FROM A SINGLE CAMERA USING INTERVAL PARTICLE FILTERING

2007 ◽  
Vol 16 (04) ◽  
pp. 593-609 ◽  
Author(s):  
JAMAL SABOUNE ◽  
FRANÇOIS CHARPILLET
Keyword(s):  
Motion Capture ◽  
Particle Filtering ◽  
Motion Tracking ◽  
Human Movement ◽  
Search Space ◽  
Human Motion ◽  
The Body ◽  
Single Camera ◽  
Human Motion Capture ◽  
Key Points

In this paper we present a new approach for marker less human motion capture from conventional camera feeds. The aim of our study is to recover 3D positions of key points of the body that can serve for gait analysis. Our approach is based on foreground extraction, an articulated body model and particle filters. In order to be generic and simple, no restrictive dynamic modeling was used. A new modified particle-filtering algorithm was introduced. It is used efficiently to search the model configurations space. This new algorithm, which we call Interval Particle Filtering, reorganizes the configurations search space in an optimal deterministic way and proved to be efficient in tracking natural human movement. Results for human motion capture from a single camera are presented and compared to results obtained from a marker based system. The system proved to be able to track motion successfully even in partial occlusions and even outdoors.

scholarly journals Movement Estimation Using Soft Sensors Based on Bi-LSTM and Two-Layer LSTM for Human Motion Capture

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1801 ◽  
Author(s):  
Haitao Guo ◽  
Yunsick Sung
Keyword(s):  
Motion Capture ◽  
Short Term Memory ◽  
Arm Movements ◽  
Ground Truth ◽  
Human Movement ◽  
Human Motion ◽  
Soft Sensors ◽  
Upper Arm ◽  
Human Motion Capture ◽  
Gesture Control

The importance of estimating human movement has increased in the field of human motion capture. HTC VIVE is a popular device that provides a convenient way of capturing human motions using several sensors. Recently, the motion of only users’ hands has been captured, thereby greatly reducing the range of motion captured. This paper proposes a framework to estimate single-arm orientations using soft sensors mainly by combining a Bi-long short-term memory (Bi-LSTM) and two-layer LSTM. Positions of the two hands are measured using an HTC VIVE set, and the orientations of a single arm, including its corresponding upper arm and forearm, are estimated using the proposed framework based on the estimated positions of the two hands. Given that the proposed framework is meant for a single arm, if orientations of two arms are required to be estimated, the estimations are performed twice. To obtain the ground truth of the orientations of single-arm movements, two Myo gesture-control sensory armbands are employed on the single arm: one for the upper arm and the other for the forearm. The proposed framework analyzed the contextual features of consecutive sensory arm movements, which provides an efficient way to improve the accuracy of arm movement estimation. In comparison with the ground truth, the proposed method estimated the arm movements using a dynamic time warping distance, which was the average of 73.90% less than that of a conventional Bayesian framework. The distinct feature of our proposed framework is that the number of sensors attached to end-users is reduced. Additionally, with the use of our framework, the arm orientations can be estimated with any soft sensor, and good accuracy of the estimations can be ensured. Another contribution is the suggestion of the combination of the Bi-LSTM and two-layer LSTM.

scholarly journals Human Motion Tracking with Less Constraint of Initial Posture from a Single RGB-D Sensor

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3029
Author(s):  
Chen Liu ◽  
Anna Wang ◽  
Chunguang Bu ◽  
Wenhui Wang ◽  
Haijing Sun
Keyword(s):  
Motion Capture ◽  
Motion Tracking ◽  
Human Motion ◽  
High Fidelity ◽  
High Quality ◽  
Front View ◽  
Human Motion Capture ◽  
Performance Capture ◽  
And Performance ◽  
4D Reconstruction

High-quality and complete human motion 4D reconstruction is of great significance for immersive VR and even human operation. However, it has inevitable self-scanning constraints, and tracking under monocular settings also has strict restrictions. In this paper, we propose a human motion capture system combined with human priors and performance capture that only uses a single RGB-D sensor. To break the self-scanning constraint, we generated a complete mesh only using the front view input to initialize the geometric capture. In order to construct a correct warping field, most previous methods initialize their systems in a strict way. To maintain high fidelity while increasing the easiness of the system, we updated the model while capturing motion. Additionally, we blended in human priors in order to improve the reliability of model warping. Extensive experiments demonstrated that our method can be used more comfortably while maintaining credible geometric warping and remaining free of self-scanning constraints.

Video-Based Motion Capture for Measuring Human Movement

2011 ◽  
pp. 81-100 ◽  
Author(s):  
Chee Kwang Quah ◽  
Michael Koh ◽  
Alex Ong ◽  
Hock Soon Seah ◽  
Andre Gagalowicz
Keyword(s):  
Computer Vision ◽  
Digital Media ◽  
Motion Capture ◽  
Human Movement ◽  
Human Motion ◽  
The Body ◽  
Optical Systems ◽  
Human Motion Analysis ◽  
Capture Method ◽  
The Subject

Through the advancement of electronics technologies, human motion analysis applications span many domains. Existing commercially available magnetic, mechanical and optical systems for motion capture and analyses are far from being able to operate in natural scenarios and environments. The current shortcoming of requiring the subject to wear sensors and markers on the body has prompted development directed towards a marker-less setup using computer vision approaches. However, there are still many challenges and problems in computer vision methods such as inconsistency of illumination, occlusion and lack of understanding and representation of its operating scenario. The authors present a videobased marker-less motion capture method that has the potential to operate in natural scenarios such as occlusive and cluttered scenes. In specific applications in sports biomechanics and education, which are stimulated by the usage of interactive digital media and augmented reality, accurate and reliable capture of human motion are essential.

Video-Based Motion Capture for Measuring Human Movement

2011 ◽  
pp. 209-228
Author(s):  
Chee Kwang Quah ◽  
Michael Koh ◽  
Alex Ong ◽  
Hock Soon Seah ◽  
Andre Gagalowicz
Keyword(s):  
Computer Vision ◽  
Digital Media ◽  
Motion Capture ◽  
Human Movement ◽  
Human Motion ◽  
The Body ◽  
Optical Systems ◽  
Human Motion Analysis ◽  
Capture Method ◽  
The Subject

Through the advancement of electronics technologies, human motion analysis applications span many domains. Existing commercially available magnetic, mechanical and optical systems for motion capture and analyses are far from being able to operate in natural scenarios and environments. The current shortcoming of requiring the subject to wear sensors and markers on the body has prompted development directed towards a marker-less setup using computer vision approaches. However, there are still many challenges and problems in computer vision methods such as inconsistency of illumination, occlusion and lack of understanding and representation of its operating scenario. The authors present a videobased marker-less motion capture method that has the potential to operate in natural scenarios such as occlusive and cluttered scenes. In specific applications in sports biomechanics and education, which are stimulated by the usage of interactive digital media and augmented reality, accurate and reliable capture of human motion are essential.

Key Points of Human Body Location Based on Single Depth Map

2012 ◽  
Vol 203 ◽  
pp. 76-82
Author(s):  
Hai Hu ◽  
Bin Li ◽  
Ben Xiong Huang ◽  
Xiao Lei He
Keyword(s):  
Human Body ◽  
Motion Tracking ◽  
Tracking System ◽  
Depth Map ◽  
Human Motion ◽  
The Body ◽  
Body Region ◽  
Automatic Initialization ◽  
Key Points ◽  
Background Segmentation

This paper presents a method of using single depth map to locate the key points of frontal human body. Human motion capture is the premise of motion analysis and understanding, and it has widely application prospects. There are many problems on former way to capture the state of human motion. For example, it can’t initialize automatically, it can not recover from tracking failure, it can not solve the problem caused by occlusion, or there are many constraints on participant, and so on. This article uses Kinect, which from Microsoft, to get depth maps, and use a single map as input to locate the key points of human body. First, depth map can reflect the distance, so background segmentation can be done easily by the characteristic. Then, extract the skeleton of the body’s silhouette. Finally, using the inherent connectivity features of human body, the key points of the body can be determined on the skeleton. Locating the key points from single depth map solve the problem of automatic initialization and recovery directly. The depth map can reflect distance on grayscale, which makes it easy to split the body region from the background. In addition, depth map contains some useful information can be used to solve the problem of occlusion. Using depth map can remove some constraints on the human body, as well as to reduce the influence of clothing and surround lighting, and so on. The experiment shows that this method is very accurate in locating the key points of frontal stand human body, and can solve some problems of occlusion. It is ideal used in a motion tracking system for automatic initialization and self-recovery when tracking failed

Sign in / Sign up

Export Citation Format

Share Document