mlsanityData to 14 Oct 2026

Real-Time Human Action Recognition on Embedded Platforms

arXiv 2409.05662 ↗First posted 9 Sep 2024cs.CV cs.AI cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

With advancements in computer vision and deep learning, video-based human action recognition (HAR) has become practical. However, due to the complexity of the computation pipeline, running HAR on live video streams incurs excessive delays on embedded platforms. This work tackles the real-time performance challenges of HAR with four contributions: 1) an experimental study identifying a standard Optical Flow (OF) extraction technique as the latency bottleneck in a state-of-the-art HAR pipeline, 2) an exploration of the latency-accuracy tradeoff between the standard and deep learning approaches to OF extraction, which highlights the need for a novel, efficient motion feature extractor, 3) the design of Integrated Motion Feature Extractor (IMFE), a novel single-shot neural network architecture for motion feature extraction with drastic improvement in latency, 4) the development of RT-HARE, a real-time HAR system tailored for embedded platforms. Experimental results on an Nvidia Jetson Xavier NX platform demonstrated that RT-HARE realizes real-time HAR at a video frame rate of 30 frames per second while delivering high levels of recognition accuracy.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 38 of its 71 references that are in the map. First posted on arXiv 9 Sep 2024.

Builds on

38 of 71 references.

References · 38

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  6. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  7. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
  8. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  9. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  10. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  11. 2019Searching for MobileNetV3cited by 1,994
  12. 2014Two-Stream Convolutional Networks for Action Recognition in Videoscited by 1,068
26 more
  1. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  2. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 856
  3. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 784
  4. 2016Hollywood in Homes: Crowdsourcing Data Collection for Activity Understandingcited by 537
  5. 2016Temporal Convolutional Networks for Action Segmentation and Detectioncited by 508
  6. 2019MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentationcited by 373
  7. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
  8. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 224
  9. 2021ASFormer: Transformer for Action Segmentationcited by 168
  10. 2021Long Short-Term Transformer for Online Action Detectioncited by 88
  11. 2020Alleviating Over-segmentation Errors by Detecting Action Boundariescited by 87
  12. 2022Temporal Action Segmentation: An Analysis of Modern Techniquescited by 85
  13. 2021OadTR: Online Action Detection with Transformerscited by 83
  14. 2022CRAFT: Cross-Attentional Flow Transformer for Robust Optical Flowcited by 82
  15. 2018Temporal Recurrent Networks for Online Action Detectioncited by 80
  16. 2016Segmental Spatiotemporal CNNs for Fine-grained Action Segmentationcited by 78
  17. 2017RED: Reinforced Encoder-Decoder Networks for Action Anticipationcited by 78
  18. 2022Global Matching with Overlapping Attention for Optical Flow Estimationcited by 64
  19. 2022Real-time Online Video Detection with Temporal Smoothing Transformerscited by 58
  20. 2024MemFlow: Optical Flow Estimation and Prediction with Memorycited by 55
  21. 2019Learning to Discriminate Information for Online Action Detectioncited by 37
  22. 2023RT-LM: Uncertainty-Aware Resource Management for Real-Time Inference of Language Modelscited by 11
  23. 2022Demand Layering for Real-Time DNN Inference with Minimized Memory Usagecited by 7
  24. 2023RED: A Systematic Real-Time Scheduling Approach for Robotic Environmental Dynamicscited by 6
  25. 2017ActivityNet Challenge 2017 Summarycited by 4
  26. 2022Two-Stream Transformer Architecture for Long Video Understandingcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025SHARDeg: A Benchmark for Skeletal Human Action Recognition in Degraded Scenarioscited by 0
  2. 2025Label-Efficient Skeleton-based Recognition with Stable-Invertible Graph Convolutional Networkscited by 0
  3. 2025Active Learning for GCN-based Action Recognitioncited by 0
  4. 2026Resource-Efficient RGB-Only Action Recognition for Edge Deploymentcited by 0