Filters








796 Hits in 4.5 sec

Feedback Graph Convolutional Network for Skeleton-based Action Recognition [article]

Hao Yang, Dan Yan, Li Zhang, Dong Li, YunDa Sun, ShaoDi You, Stephen J. Maybank
2020 arXiv   pre-print
In this paper, we propose a novel network, named Feedback Graph Convolutional Network (FGCN). This is the first work that introduces the feedback mechanism into GCNs and action recognition.  ...  Skeleton-based action recognition has attracted considerable attention in computer vision since skeleton data is more robust to the dynamic circumstance and complicated background than other modalities  ...  GCN based Action Recognition The Graph Convolutional Networks (GCNs) [2, 34, 7, 12, 19] generalize the convolutional operation to deal with the data with graph construction.  ... 
arXiv:2003.07564v1 fatcat:mp7qkpj6hzbnzprdtrnvl2khj4

Whole and Part Adaptive Fusion Graph Convolutional Networks for Skeleton-Based Action Recognition

Qi Zuo, Lian Zou, Cien Fan, Dongqian Li, Hao Jiang, Yifeng Liu
2020 Sensors  
Spatiotemporal graph convolution has made significant progress in skeleton-based action recognition in recent years.  ...  and the part graph convolution network (PGCN).  ...  We propose two new graph convolution methods for skeleton-based action recognition based on the movement patterns of human action: whole graph convolution (WGCN) and part graph convolution (PGCN), which  ... 
doi:10.3390/s20247149 pmid:33322231 fatcat:dgvzytru6rdfjb4ypbqhc3nika

A Lightweight Hierarchical Model with Frame-Level Joints Adaptive Graph Convolution for Skeleton-Based Action Recognition

Yujian Jiang, Xue Yang, Jingyu Liu, Junming Zhang, Zhenhua Tan
2021 Security and Communication Networks  
To obtain a model with fewer parameters and a higher accuracy, this study designed a lightweight frame-level joints adaptive graph convolutional network (FLAGCN) model to solve skeleton-based action recognition  ...  In skeleton-based human action recognition methods, human behaviours can be analysed through temporal and spatial changes in the human skeleton.  ...  Acknowledgments is work was supported by Funds for Key Laboratory of Ministry of Culture and Tourism (WLBSYS2005) and the Fundamental Research Funds for the Central Universities (CUC19ZD005).  ... 
doi:10.1155/2021/2290304 fatcat:cxtjpygsqffcxi7xmlhpea2tmu

Action Recognition with Fusion of Multiple Graph Convolutional Networks

Camille Maurice, Frederic Lerasle
2021 2021 17th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)  
We propose two light-weight and specialized Spatio-Temporal Graph Convolutional Networks (ST-GCNs): one for actions characterized by the motion of the human body and a novel one we especially design to  ...  We propose a late-fusion strategy of the predictions of both graphs networks to get the most out of the two and to clear out ambiguities in the action classification.  ...  Especially in the action recognition topic where the human skeleton joints and limbs naturally define a graph. We propose a modular framework for action recognition based on GCNs.  ... 
doi:10.1109/avss52988.2021.9663765 fatcat:m65zkcw2xjeq3db2yihbrl3osa

Fusion-GCN: Multimodal Action Recognition using Graph Convolutional Networks [article]

Michael Duhme, Raphael Memmesheimer, Dietrich Paulus
2021 arXiv   pre-print
In this paper, we present Fusion-GCN, an approach for multimodal action recognition using Graph Convolutional Networks (GCNs).  ...  Action recognition methods based around GCNs recently yielded state-of-the-art performance for skeleton-based action recognition.  ...  existing GCN models for skeleton-based action recognition through data fusion and augmentation of skeleton sequences.  ... 
arXiv:2109.12946v1 fatcat:3z5t4er56vf2zhiwxpvl7o5f6a

Skeleton Motion Recognition Based on Multi-Scale Deep Spatio-Temporal Features

Kai Hu, Yiwu Ding, Junlan Jin, Liguo Weng, Min Xia
2022 Applied Sciences  
We study and compare the performance of three existing multi-channel fusion methods to improve the recognition accuracy of the network on the open skeleton recognition dataset.  ...  In the task of human motion recognition, the overall action span is changeable, and there may be an inclusion relationship between action semantics.  ...  With the rapid development of graph convolutional neural networks (GCNs), a large number of graph convolution-based network models have been applied to skeleton-based action recognition tasks.  ... 
doi:10.3390/app12031028 fatcat:qc3adyv62jemlewzygjwc5k6wa

RGB-D Data-Based Action Recognition: A Review

Muhammad Bilal Shaikh, Douglas Chai
2021 Sensors  
Naturally, each action-data modality—such as RGB, depth, skeleton, and infrared (IR)—has distinct characteristics; therefore, it is important to exploit the value of each modality for better action recognition  ...  This review is aimed to scope current literature on data fusion and action recognition techniques and to identify gaps and future research direction.  ...  Acknowledgments: The authors would like to thank the anonymous reviewers for their careful reading and valuable remarks, which have greatly helped extend the scope of this paper.  ... 
doi:10.3390/s21124246 fatcat:7dvocdy63rckne5yunhfsnr4p4

Multi-Stage Attention-Enhanced Sparse Graph Convolutional Network for Skeleton-Based Action Recognition

Chaoyue Li, Lian Zou, Cien Fan, Hao Jiang, Yifeng Liu
2021 Electronics  
To better extract spatial-temporal features, we propose a multi-stage attention-enhanced sparse graph convolutional network (MS-ASGCN) for skeleton-based action recognition.  ...  Graph convolutional networks (GCNs), which model human actions as a series of spatial-temporal graphs, have recently achieved superior performance in skeleton-based action recognition.  ...  Network architecture of MS-ASGCN. Table 1 . 1 Pros and cons of skeleton-based action recognition methods.  ... 
doi:10.3390/electronics10182198 fatcat:nx2o7sckrrcrfpnk3wcrzzmw3u

Multi‐stream adaptive spatial‐temporal attention graph convolutional network for skeleton‐based action recognition

Lubin Yu, Lianfang Tian, Qiliang Du, Jameel Ahmed Bhutto
2021 IET Computer Vision  
Graph convolutional networks (GCNs) generalize convolutional neural networks (CNNs) to non-Euclidean graphs and achieve significant performance in skeleton-based action recognition.  ...  Skeleton-based action recognition algorithms have been widely applied to human action recognition.  ...  [12] used multiscale residual networks and several data enhancement strategies for skeleton-based action recognition.  ... 
doi:10.1049/cvi2.12075 fatcat:auyz7ymr3fevhphfk32qzwmqae

A Two-Stream Recurrent Network for Skeleton-based Human Interaction Recognition

Qianhui Men, Edmond S. L. Ho, Hubert P. H. Shum, Howard Leung
2021 2020 25th International Conference on Pattern Recognition (ICPR)  
joints, where the joint connections in the graph are adaptive for flexible correlations.  ...  Moreover, the late fusion results demonstrate the effectiveness of improving the recognition accuracy compared with single streams.  ...  Graph convolutional network (GCN) [11] has become an effective tool for analyzing joint correlations within a single action sequence under graphical structure [12] , [13] .  ... 
doi:10.1109/icpr48806.2021.9412538 fatcat:p4bi7dqmbnggtohdar42e5gwee

Multi-scale Mixed Dense Graph Convolution Network for Skeleton-based Action Recognition

Hailun Xia, Xinkai Gao
2021 IEEE Access  
In skeleton-based action recognition, the approaches based on graph convolutional networks(GCN) have achieved remarkable performance by modeling spatial-temporal graphs to explore the physical dependencies  ...  INDEX TERMS Dense graph convolution, spatial and temporal attention module, multi-scale mixed temporal convolution, skeleton-based action recognition.  ...  Convolutional neural networks also achieve remarkable results for skeleton-based action recognition.  ... 
doi:10.1109/access.2020.3049029 fatcat:xlmmcsmp3vbnvj422wctwwjiei

Spatio-Temporal Pyramid Graph Convolutions for Human Action Recognition and Postural Assessment [article]

Behnoosh Parsa, Athma Narayanan, Behzad Dariush
2019 arXiv   pre-print
Among them, methods based on graph convolutional networks that extract features from the skeleton have demonstrated promising performance.  ...  In this paper, we propose a novel Spatio-Temporal Pyramid Graph Convolutional Network (ST-PGN) for online action recognition for ergonomic risk assessment that enables the use of features from all levels  ...  Most related work in skeleton based action recognition include [58, 17, 24, 47] . The first three papers focus on graph convolution on temporal skeleton sequences.  ... 
arXiv:1912.03442v1 fatcat:k2hmpiydezagdgsvcwd6niirhe

Robust Multi-Feature Learning for Skeleton-Based Action Recognition

Yingfu Wang, Zheyuan Xu, Li Li, Yao Jian
2019 IEEE Access  
Skeleton-based action recognition has advanced significantly in the past decade.  ...  Among deep learning-based action recognition methods, one of the most commonly used structures is a two-stream network.  ...  CONCLUSION In this paper, we propose a fully convolutional network for skeleton-based action recognition that can robustly learn and fuse multiple low-level skeleton features.  ... 
doi:10.1109/access.2019.2945632 fatcat:6c4i2axfnff5rin4ckhjjitwbe

Pyramid Spatial-Temporal Graph Transformer for Skeleton-Based Action Recognition

Shuo Chen, Ke Xu, Xinghao Jiang, Tanfeng Sun
2022 Applied Sciences  
Although graph convolutional networks (GCNs) have shown their demonstrated ability in skeleton-based action recognition, both the spatial and the temporal connections rely too much on the predefined skeleton  ...  graph, which imposes a fixed prior knowledge for the aggregation of high-level semantic information via the graph-based convolution.  ...  [10] first proposed a spatial-temporal graph convolutional network (ST-GCN) and provided a robust GCN backbone for skeleton-based action recognition.  ... 
doi:10.3390/app12189229 fatcat:rcbcxavdn5edxlyzhybvyetsjm

Revisiting Skeleton-based Action Recognition [article]

Haodong Duan, Yue Zhao, Kai Chen, Dahua Lin, Bo Dai
2022 arXiv   pre-print
Many skeleton-based action recognition methods adopt graph convolutional networks (GCN) to extract features on top of human skeletons.  ...  In this work, we propose PoseC3D, a new approach to skeleton-based action recognition, which relies on a 3D heatmap stack instead of a graph sequence as the base representation of human skeletons.  ...  Graph convolutional network is widely adopted in skeleton-based action recognition [3, 7, 19, 55, 56, 71] . It models human skeleton sequences as spatiotemporal graphs.  ... 
arXiv:2104.13586v2 fatcat:b64skl5xtffy3c22awbdhgxh3m
« Previous Showing results 1 — 15 out of 796 results