An Image-Constrained Particle Filter for 3D Human Motion Tracking

Zhao, Xiukai; Lyu, Lei; Zhang, Jinling; Lyu, Chen

doi:10.1109/access.2019.2891172

Cited by 4 publications

(2 citation statements)

References 38 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many motion capture techniques have been proposed based on different methods. Human kinematic measurements are discussed in many literature surveys, such as [6], [7] while performing motion capture from a single image is proposed in [8]- [12]. As different human pose estimations cannot capture the same silhouette using a single camera, this led researchers to use multiple cameras.…”

Section: Related Workmentioning

confidence: 99%

Stereo Vision-Based 3D Positioning and Tracking

et al. 2020

View full text Add to dashboard Cite

The evolution of technologies for the capture of human movement has been motivated by a number of potential applications across a wide variety of fields. However, capturing human motion in 3D is difficult in an outdoor environment when it is performed without controlled surroundings. In this paper, a stereo camera rig with an ultra-wide baseline distance and conventional cameras with fish-eye lenses is proposed. Its cameras provide a wide field of view (FOV) which increases the coverage area and also enables the baseline distance to be increased to cover the common area required for both cameras' views to perform as a stereo camera. We propose a passive marker-based approach to track the motion of the object. In this method, an adaptive thresholding method is applied to extract each small pink polyester marker from the video frames. As the cameras have fish-eye lenses, it is difficult to estimate the depth information using a pinhole camera model. We use a unique method to restore the 3D positions by developing a relationship between the pixel dimensions and distances in an image and real world coordinates. In this paper, occlusion detection is considered because, in the marker-based capturing of articulated human kinematics, the occlusion of a marker is one of the major challenges. The detection algorithm differentiates among types of occlusions and predicts any missing marker position where necessary. As this design is intended to be mounted on a moving carrier, such as a drone or car, a method for compensating the camera's ego-motion is proposed. The proposed 3D positioning and tracking system is tested in different situations to validate its applicability as a stereo camera rig as well as its performance for motion capture. The performance of the proposed system is compared with that of a standard motion capture system called Vicon and is shown to have the same order of accuracy while incurring less cost. INDEX TERMS Motion capture, 3D positioning, stereo vision, motion tracking, high precision.

show abstract

Section: Related Workmentioning

confidence: 99%

Stereo Vision-Based 3D Positioning and Tracking

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Additionally, the concept of feature fusion in the particle filter framework has been adopted by some researchers to address the specific issues of the scene. The issue of 3D-Human motion tracking is carried out by modeling the target with a fusion of contour and edge features [41]. Tracking in a cluttered environment is achieved by the fusion of shape and texture features [42].…”

Section: Introductionmentioning

confidence: 99%

Particle Filter Based Video Object Tracking Scheme With Target Remodeling and Reinitialization and its Hardware Implementation Using Raspberry Pi

Panda

Nanda

Pradhan

2023

Preprint

View full text Add to dashboard Cite

Video object tracking in real-world scenarios is one of the challenging problems of computer vision. The issue is compounded in the presence of varying illumination conditions, dynamic entities of the background, and bad weather conditions. In this research, a particle filter based new video object tracking scheme is developed with the proposed notions of target remodeling and reinitialization. During the tracking phase, the target is remodeled in each frame to take care of the changing scene dynamics over frames. The target is remodeled by fused feature distributions chosen from the created bank of fused feature distributions having discriminating potential to differentiate the target and the background in a given frame. The fused feature bank is created by fusing two features from the set consisting of Color, LBP, and HOG features. The features are fused probabilistically where the weights are determined based on the discriminating ability of a given feature. In order to achieve high tracking accuracy, the deviation of the tracker is evaluated ineach frame using the notion of time motion history and reinitialization of the tracker position takes place when the deviation is above a preselected threshold. Besides, the proposed algorithm has been implemented successfully on a Raspberry Pi based hardware setup and thus becomes a potential candidate for real time implementation. The proposed scheme is successfully tested on videos from DAVIS 2016, LASIESTA, and CDnet 2014 databases and the tracking accuracy is found to be higher than those of the existing algorithms.

show abstract

Particle filter-based video object tracking using feature fusion in template partitions

Panda

Nanda

2022

Vis Comput

View full text Add to dashboard Cite

An Image-Constrained Particle Filter for 3D Human Motion Tracking

Cited by 4 publications

References 38 publications

Stereo Vision-Based 3D Positioning and Tracking

Stereo Vision-Based 3D Positioning and Tracking

Particle Filter Based Video Object Tracking Scheme With Target Remodeling and Reinitialization and its Hardware Implementation Using Raspberry Pi

Particle filter-based video object tracking using feature fusion in template partitions

Contact Info

Product

Resources

About