Pose-Guided Tracking-by-Detection: Robust Multi-Person Pose Tracking

Bao, Qing; Liu, Wu; Cheng, Yuhao; Zhou, Boyan; Mei, Tao

doi:10.1109/tmm.2020.2980194

Cited by 62 publications

(21 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There are prior works that use GNNs for generic object tracking [13,2]. Gao et al [13] proposed to divide an object into several parts and learn a spatial-temporal template of the object for tracking.…”

Section: Graph Neural Networkmentioning

confidence: 99%

“…Gao et al [13] proposed to divide an object into several parts and learn a spatial-temporal template of the object for tracking. Bao et al [2] utilized GNN in their pose tracking method to exploit human structural relations to help associate human poses across frames. This method relies on a strong human detector as well as a strong pose estimator to generate human poses for association.…”

Section: Graph Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking

Yang¹,

Ren²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

Multi-person pose estimation and tracking serve as crucial steps for video understanding. Most state-of-the-art approaches rely on first estimating poses in each frame and only then implementing data association and refinement. Despite the promising results achieved, such a strategy is inevitably prone to missed detections especially in heavilycluttered scenes, since this tracking-by-detection paradigm is, by nature, largely dependent on visual evidences that are absent in the case of occlusion. In this paper, we propose a novel online approach to learning the pose dynamics, which are independent of pose detections in current fame, and hence may serve as a robust estimation even in challenging scenarios including occlusion. Specifically, we derive this prediction of dynamics through a graph neural network (GNN) that explicitly accounts for both spatialtemporal and visual information. It takes as input the historical pose tracklets and directly predicts the corresponding poses in the following frame for each tracklet. The predicted poses will then be aggregated with the detected poses, if any, at the same frame so as to produce the final pose, potentially recovering the occluded joints missed by the estimator. Experiments on PoseTrack 2017 and Pose-Track 2018 datasets demonstrate that the proposed method achieves results superior to the state of the art on both human pose estimation and tracking tasks.

show abstract

Section: Graph Neural Networkmentioning

confidence: 99%

Section: Graph Neural Networkmentioning

confidence: 99%

Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking

Yang¹,

Ren²,

Li³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Jin et al (2019) proposed SpatialNet and TemporalNet combined to form a single pose prediction and monitoring conceptual model: Body part identification and part-level temporal classification are handled by SpatialNet, while the contextual classification of human events is handled by TemporalNet. Bao et al (2020) suggest a hand gesture identification-by-tracking system that incorporates pose input into both the video human identification and human connection levels. A person's position prediction with pose descriptive statistics is used in the first level to reduce the impact of distracting and incomplete human identification in images.…”

Section: Experiments I: the Landmarks Detection Accuraciesmentioning

confidence: 99%

Peer Review #1 of "Syntactic model-based human body 3D reconstruction and event classification via association based features mining and deep learning (v0.1)"

Nayyar¹

2021

View full text Add to dashboard Cite

“…In addition, a local-to-global strategy for robust data association has been introduced in [12]. In recent years, some dedicated tracking algorithms have been designed either to increase the recall and accuracy of the detector, or to enhance the performance of data association [13]- [18].…”

Section: A Mot With Machine Learningmentioning

confidence: 99%

A CRF-based Framework for Tracklet Inactivation in Online Multi-Object Tracking

Gao,

Pan,

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

Online multi-object tracking (MOT) is an active research topic in the domain of computer vision. In this paper, a CRF-based framework is put forward to tackle the tracklet inactivation issues in online MOT problems. We apply the proposed framework to one of the state-of-the-art online MOT trackers, Tracktor++. The baseline algorithm for online MOT has the drawback of simple strategy on tracklet inactivation, which relies merely on tracking hypotheses' classification scores partitioned by using a fixed threshold. To overcome such a drawback, a discrete conditional random field (CRF) is developed to exploit the intra-frame relationship between tracking hypotheses. Separate sets of feature functions are designed for the unary and binary terms in the CRF so as to cope with various challenges in practical situations. The hypothesis filtering and dummy nodes techniques are employed to handle the problem of varying CRF nodes in the MOT context. In this paper, the inference of CRF is achieved by using the loopy belief propagation algorithm, and the parameters of the CRF are determined by utilizing the maximum likelihood estimation method. Experimental results demonstrate that the developed tracker with our CRF-based framework outperforms the baseline on the MOT16 and MOT17 datasets. The extensibility of the proposed method is further validated by an extensive experiment.

show abstract

Pose-Guided Tracking-by-Detection: Robust Multi-Person Pose Tracking

Cited by 62 publications

References 48 publications

Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking

Learning Dynamics via Graph Neural Networks for Human Pose Estimation and Tracking

Peer Review #1 of "Syntactic model-based human body 3D reconstruction and event classification via association based features mining and deep learning (v0.1)"

A CRF-based Framework for Tracklet Inactivation in Online Multi-Object Tracking

Contact Info

Product

Resources

About