211 Hits in 4.3 sec

Dynamic Multiscale Graph Neural Networks for 3D Skeleton-Based Human Motion Prediction [article]

Maosen Li, Siheng Chen, Yangheng Zhao, Ya Zhang, Yanfeng Wang, Qi Tian
2020 arXiv   pre-print
We propose novel dynamic multiscale graph neural networks (DMGNN) to predict 3D skeleton-based human motions.  ...  This multiscale graph is adaptive during training and dynamic across network layers.  ...  Conclusion We build dynamic mutiscale graphs to represent a human body and propose dynamic multiscale graph neural networks (DMGNN) with an encoder-decoder framework for 3D skeleton-based human motion  ... 
arXiv:2003.08802v1 fatcat:pexi6av5d5fdrjbg7ywh5jsyjy

Learning Multiscale Correlations for Human Motion Prediction [article]

Honghong Zhou, Caili Guo, Hao Zhang, Yanjun Wang
2021 arXiv   pre-print
In this paper, we propose a novel multiscale graph convolution network (MGCN) to address this problem.  ...  We evaluate our approach on two standard benchmark datasets for human motion prediction: Human3.6M and CMU motion capture dataset.  ...  INTRODUCTION Human motion prediction aims to use the 3D skeleton data to predict a sequence of future human motions based on observed motion frames.  ... 
arXiv:2103.10674v2 fatcat:abb4fn4qcrce5ig3tubhfu36cu

Symbiotic Graph Neural Networks for 3D Skeleton-based Human Action Recognition and Motion Prediction [article]

Maosen Li, Siheng Chen, Xu Chen, Ya Zhang, Yanfeng Wang, Qi Tian
2019 arXiv   pre-print
3D skeleton-based action recognition and motion prediction are two essential problems of human activity understanding.  ...  We conduct extensive experiments for skeleton-based action recognition and motion prediction with four datasets, NTU-RGB+D, Kinetics, Human3.6M, and CMU Mocap.  ...  We propose novel symbiotic graph neural networks (Sym-GNN) to achieve 3D skeleton-based action recognition and motion prediction in a multitasking framework.  ... 
arXiv:1910.02212v1 fatcat:suouaghedffsfaszejnp375yxy

DMS-GCN: Dynamic Mutiscale Spatiotemporal Graph Convolutional Networks for Human Motion Prediction [article]

Zigeng Yan, Di-Hua Zhai, Yuanqing Xia
2021 arXiv   pre-print
In this paper, we propose a simple feed-forward deep neural network for motion prediction, which takes into account temporal smoothness and spatial dependencies between human body joints.  ...  We design a Multi-scale Spatio-temporal graph convolutional networks (GCNs) to implicitly establish the Spatio-temporal dependence in the process of human movement, where different scales fused dynamically  ...  Conclusion and Future Work In this paper, we build dynamic multiscale spatiotemporal graph convolutional networks to effectively predict future human poses from observed histories.  ... 
arXiv:2112.10365v1 fatcat:m6vaxivreba2xha6ffcp32ddqe

ANUBIS: Skeleton Action Recognition Dataset, Review, and Benchmark [article]

Zhenyue Qin, Yang Liu, Madhawa Perera, Tom Gedeon, Pan Ji, Dongwoo Kim, Saeed Anwar
2022 arXiv   pre-print
We aim to provide a roadmap for new and existing researchers a on the landscapes of skeleton-based action recognition for new and existing researchers.  ...  Compared with other data modalities, 3D human skeleton representations have extensive unique desirable characteristics, including succinctness, robustness, racial-impartiality, and many more.  ...  and graph neural networks, in order to effectively capture the complex spatial-temporal movement patterns of human skeletons.  ... 
arXiv:2205.02071v3 fatcat:22iwew3p4faqxfp2ozbehyeffu

Motion Guided 3D Pose Estimation from Videos [article]

Jingbo Wang, Sijie Yan, Yuanjun Xiong, Dahua Lin
2020 arXiv   pre-print
We propose a new loss function, called motion loss, for the problem of monocular 3D Human pose estimation from 2D pose.  ...  We design a new graph convolutional network architecture, U-shaped GCN (UGCN).  ...  We structure 2D skeletons by a spatial-temporal graph and predict 3D locations via our U-shaped Graph Convolution Networks (UGCN).  ... 
arXiv:2004.13985v1 fatcat:7vnvq3pp4bdhjdfqn6tj4zhg54

Fusing Higher-Order Features in Graph Neural Networks for Skeleton-Based Action Recognition [article]

Zhenyue Qin and Yang Liu and Pan Ji and Dongwoo Kim and Lei Wang and Bob McKay and Saeed Anwar and Tom Gedeon
2021 arXiv   pre-print
Recent skeleton-based action recognition methods extract features from 3D joint coordinates as spatial-temporal cues, using these representations in a graph neural network for feature fusion to boost recognition  ...  This simple fusion with popular spatial-temporal graph neural networks achieves new state-of-the-art accuracy in two large benchmarks, including NTU60 and NTU120, while employing fewer parameters and reduced  ...  Graph neural networks started to attract attention [18] , [13] , [30] in skeleton recognition. In GCN-based models, a skeleton is treated as a graph, with joints as nodes and bones as edges.  ... 
arXiv:2105.01563v4 fatcat:oxzgap7efraqznhsqkkzl2rmwy

Action Recognition Using Action Sequences Optimization and Two-Stream 3D Dilated Neural Network

Xin Xiong, Weidong Min, Qing Han, Qi Wang, Cheng Zha, Hubert Cecotti
2022 Computational Intelligence and Neuroscience  
A two-stream 3D dilated neural network that integrates features of RGB and human skeleton information is also proposed.  ...  The human skeleton information strengthens the deep representation of humans for robust processing, alleviating the interference of background changes, and the dilated CNN enlarges the receptive field  ...  temporal information. (2) A two-stream 3D dilated convolution neural network integrates features of RGB and human skeleton information is also proposed. e human skeleton information strengthens the deep  ... 
doi:10.1155/2022/6608448 fatcat:m56ufg2z25goxdqzphmbsxaili

SM-SGE: A Self-Supervised Multi-Scale Skeleton Graph Encoding Framework for Person Re-Identification [article]

Haocong Rao, Xiping Hu, Jun Cheng, Bin Hu
2021 arXiv   pre-print
In this paper, we for the first time propose a Self-supervised Multi-scale Skeleton Graph Encoding (SM-SGE) framework that comprehensively models human body, component relations, and skeleton dynamics  ...  Specifically, we first devise multi-scale skeleton graphs with coarse-to-fine human body partitions, which enables us to model body structure and skeleton dynamics at multiple levels.  ...  [16] propose PoseGait model, which learns 81 hand-crafted pose features of 3D skeleton data with deep convolutional neural networks (CNN) for gait-based human recognition.  ... 
arXiv:2107.01903v1 fatcat:5mduyl2xdjepxhyxlz43uo2ixe

Multi-Scale Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognition [article]

Pengfei Zhang and Cuiling Lan and Wenjun Zeng and Junliang Xing and Jianru Xue and Nanning Zheng
2021 arXiv   pre-print
In this paper, a simple yet effective multi-scale semantics-guided neural network (MS-SGN) is proposed for skeleton-based action recognition.  ...  However, there is a trend of using very deep and complicated feedforward neural networks to model the skeleton sequence without considering the complexity in recent year.  ...  Moreover, the scalable motion dynamics are in general under-explored in skeleton based action recognition.  ... 
arXiv:2111.03993v1 fatcat:v4pfwjjwhfbpvcijacqhpm3xxe

A Survey of Vision-Based Human Action Evaluation Methods

Qing Lei, Ji-Xiang Du, Hong-Bo Zhang, Shuang Ye, Duan-Sheng Chen
2019 Sensors  
This paper presents a comprehensive survey of approaches and techniques in action evaluation research, including motion detection and preprocessing using skeleton data, handcrafted feature representation  ...  The research on human action evaluation differs by aiming to design computation models and evaluation approaches for automatically assessing the quality of human actions.  ...  Acknowledgments: The authors would like to thank the anonymous reviewers for their valuable and insightful comments on an earlier version of this manuscript.  ... 
doi:10.3390/s19194129 pmid:31554229 pmcid:PMC6806217 fatcat:zgwsdv6xorfvxjck6rsjusezwe

Three-Dimensional Diffusion Model in Sports Dance Video Human Skeleton Detection and Extraction

Zhi Li, Miaochao Chen
2021 Advances in Mathematical Physics  
The research in this paper mainly includes as follows: for the principle of action recognition based on the 3D diffusion model convolutional neural network, the whole detection process is carried out from  ...  graph matching, thus proposing a matching algorithm for discrete skeleton points and optimizing it for the skeleton dislocation and algorithm problems of human occlusion.  ...  Principle of Action Recognition Based on THE 3D Diffusion Model Convolutional Neural Network.  ... 
doi:10.1155/2021/3772358 fatcat:chfv5pithngubfcaqsbfwj2msm

Multi-Scale Adaptive Aggregate Graph Convolutional Network for Skeleton-Based Action Recognition

Zhiyun Zheng, Yizhou Wang, Xingjin Zhang, Junfeng Wang
2022 Applied Sciences  
In this paper, we propose a multi-scale adaptive aggregate graph convolution network (MSAAGCN) for skeleton-based action recognition.  ...  In recent years, there is a trend of using graph convolutional networks (GCNs) to model the human skeleton into a spatio-temporal graph to explore the internal connections of human joints that has achieved  ...  In traditional deep learning methods, the skeleton sequence is usually fed into recurrent neural networks (RNN) or convolutional neural networks (CNN) for analysis, and its features are captured to predict  ... 
doi:10.3390/app12031402 fatcat:3ceyzh5tjbb3jbkvx4a5nbeytm

Kinematic Motion Retargeting via Neural Latent Optimization for Learning Sign Language [article]

Haodong Zhang, Weijie Li, Jiangpin Liu, Zexi Chen, Yuxiang Cui, Yue Wang, Rong Xiong
2022 arXiv   pre-print
Traditional optimization-based methods are time-consuming and rely heavily on good initialization, while recent studies using feedforward neural networks suffer from poor generalization to unseen motions  ...  Both the human skeleton and the robot structure are modeled as graphs to make better use of topological information.  ...  to learn the significant information of human body skeletons for action recognition. [16] introduced a dynamic multiscale graph neural network to predict 3D skeleton-based human motions. [28] presented  ... 
arXiv:2103.08882v4 fatcat:khe7oz2qnzc27bv4hosyfacekm

MSR-GCN: Multi-Scale Residual Graph Convolution Networks for Human Motion Prediction [article]

Lingwei Dang, Yongwei Nie, Chengjiang Long, Qing Zhang, Guiqing Li
2021 arXiv   pre-print
Recently, graph convolutional network has been proven to be very effective to learn dynamic relations among pose joints, which is helpful for pose prediction.  ...  In this paper, we propose a novel Multi-Scale Residual Graph Convolution Network (MSR-GCN) for human pose prediction task in the manner of end-to-end.  ...  Science Foundation of China (62072191, 61802453, 61972160), in part by the Natural Science Foundation of Guangdong Province (2019A1515010860, 2021A1515012301), and in part by the Fundamental Research Funds for  ... 
arXiv:2108.07152v2 fatcat:x5ehkmkh6jcgtajbcupo6qpeya
« Previous Showing results 1 — 15 out of 211 results