mlsanityData to 7 Oct 2026

Real-World Graph Convolution Networks (RW-GCNs) for Action Recognition in Smart Video Surveillance

arXiv 2201.05739 ↗First posted 15 Jan 2022cs.CV cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Action recognition is a key algorithmic part of emerging on-the-edge smart video surveillance and security systems. Skeleton-based action recognition is an attractive approach which, instead of using RGB pixel data, relies on human pose information to classify appropriate actions. However, existing algorithms often assume ideal conditions that are not representative of real-world limitations, such as noisy input, latency requirements, and edge resource constraints. To address the limitations of existing approaches, this paper presents Real-World Graph Convolution Networks (RW-GCNs), an architecture-level solution for meeting the domain constraints of Real World Skeleton-based Action Recognition. Inspired by the presence of feedback connections in the human visual cortex, RW-GCNs leverage attentive feedback augmentation on existing near state-of-the-art (SotA) Spatial-Temporal Graph Convolution Networks (ST-GCNs). The ST-GCNs' design choices are derived from information theory-centric principles to address both the spatial and temporal noise typically encountered in end-to-end real-time and on-the-edge smart video systems. Our results demonstrate RW-GCNs' ability to serve these applications by achieving a new SotA accuracy on the NTU-RGB-D-120 dataset at 94.1%, and achieving 32X less latency than baseline ST-GCN applications while still achieving 90.4% accuracy on the Northwestern UCLA dataset in the presence of spatial keypoint noise. RW-GCNs further show system scalability by running on the 10X cost effective NVIDIA Jetson Nano (as opposed to NVIDIA Xavier NX), while still maintaining a respectful range of throughput (15.6 to 5.5 Actions per Second) on the resource constrained device. The code is available here: https://github.com/TeCSAR-UNCC/RW-GCN.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 39 of its 69 references that are in the map. First posted on arXiv 15 Jan 2022.

Builds on

39 of 69 references.

References · 39

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  4. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  5. 2017The Kinetics Human Action Video Datasetcited by 1,894
  6. 2018SlowFast Networks for Video Recognitioncited by 1,862
  7. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
  8. 2018OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,161
  9. 2017A Closer Look at Spatiotemporal Convolutions for Action Recognitioncited by 1,139
  10. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 852
  11. 2015Deep Learning and the Information Bottleneck Principlecited by 793
  12. 2019ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networkscited by 572
27 more
  1. 2019NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understandingcited by 530
  2. 2017Opening the Black Box of Deep Neural Networks via Informationcited by 494
  3. 2019Attention Augmented Convolutional Networkscited by 363
  4. 2020Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognitioncited by 339
  5. 2019On the Relationship between Self-Attention and Convolutional Layerscited by 299
  6. 2019Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognitioncited by 228
  7. 2019An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognitioncited by 151
  8. 2019STM: SpatioTemporal and Motion Encoding for Action Recognitioncited by 142
  9. 2014Cross-view Action Modeling, Learning and Recognitioncited by 134
  10. 2021Beyond Self-attention: External Attention using Two Linear Layers for Visual Taskscited by 109
  11. 2020Stronger, Faster and More Explainable: A Graph Convolutional Baseline for Skeleton-based Action Recognitioncited by 91
  12. 2018Multi-granularity Generator for Temporal Action Proposalcited by 77
  13. 2020Spatial Temporal Transformer Network for Skeleton-based Action Recognitioncited by 76
  14. 2019Learning Implicitly Recurrent CNNs Through Parameter Sharingcited by 27
  15. 2018Scalable Mutual Information Estimation using Dependence Graphscited by 22
  16. 2020Feedback Graph Convolutional Network for Skeleton-based Action Recognitioncited by 22
  17. 2020Decoupled Spatial-Temporal Attention Network for Skeleton-Based Action Recognitioncited by 21
  18. 2018Convolutional Neural Networks with Alternately Updated Cliquecited by 14
  19. 2019REVAMP2T: Real-time Edge Video Analytics for Multi-camera Privacy-aware Pedestrian Trackingcited by 14
  20. 2020EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose Estimationcited by 12
  21. 2019Information Plane Analysis of Deep Neural Networks via Matrix-Based Renyi's Entropy and Tensor Kernelscited by 11
  22. 2018Real-Time Human Detection as an Edge Service Enabled by a Lightweight CNNcited by 8
  23. 2019Vertex Feature Encoding and Hierarchical Temporal Modeling in a Spatial-Temporal Graph Convolutional Network for Action Recognitioncited by 8
  24. 2020Predictively Encoded Graph Convolutional Network for Noise-Robust Skeleton-based Action Recognitioncited by 8
  25. 2017ShaResNet: reducing residual network parameter number by sharing weightscited by 5
  26. 2020What Information Does a ResNet Compress?cited by 5
  27. 2019I-SAFE: Instant Suspicious Activity identiFication at the Edge using Fuzzy Decision Makingcited by 1

Built on by

Cited by year

Citing papers per year: 2022: 4; 2023: 0; 2024: 0; 2025: 2.

6 papers.

  1. 2022Semantic2Graph: Graph-based Multi-modal Feature Fusion for Action Segmentation in Videoscited by 9
  2. 2022Pishgu: Universal Path Prediction Network Architecture for Real-time Cyber-physical Edge Systemscited by 9
  3. 2022Understanding Ethics, Privacy, and Regulations in Smart Video Surveillance for Public Safetycited by 7
  4. 2022ADG-Pose: Automated Dataset Generation for Real-World Human Pose Estimationcited by 2
  5. 2025Action Recognition in Real-World Ambient Assisted Living Environmentcited by 2
  6. 2025Adversarially-Refined VQ-GAN with Dense Motion Tokenization for Spatio-Temporal Heatmapscited by 2