mlsanityData to 14 Oct 2026

H-MoRe: Learning Human-centric Motion Representation for Action Analysis

arXiv 2504.10676 ↗First posted 14 Apr 2025cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

In this paper, we propose H-MoRe, a novel pipeline for learning precise human-centric motion representation. Our approach dynamically preserves relevant human motion while filtering out background movement. Notably, unlike previous methods relying on fully supervised learning from synthetic data, H-MoRe learns directly from real-world scenarios in a self-supervised manner, incorporating both human pose and body shape information. Inspired by kinematics, H-MoRe represents absolute and relative movements of each body point in a matrix format that captures nuanced motion details, termed world-local flows. H-MoRe offers refined insights into human motion, which can be integrated seamlessly into various action-related applications. Experimental results demonstrate that H-MoRe brings substantial improvements across various downstream tasks, including gait recognition(CL@R1: +16.01%), action recognition(Acc@1: +8.92%), and video generation(FVD: -67.07%). Additionally, H-MoRe exhibits high inference efficiency (34 fps), making it suitable for most real-time scenarios. Models and code will be released upon publication.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 54 of its 80 references that are in the map. First posted on arXiv 14 Apr 2025.

Builds on

54 of 80 references.

References · 54

  1. 2017Mask R-CNNcited by 7,277
  2. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
  3. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  4. 2018Towards Accurate Generative Models of Video: A New Metric & Challengescited by 1,234
  5. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,209
  6. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,059
  7. 2016Temporal Segment Networks: Towards Good Practices for Deep Action Recognitioncited by 902
  8. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
  9. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 784
  10. 2023DynamiCrafter: Animating Open-domain Images with Video Diffusion Priorscited by 471
  11. 2020U2-Net: Going Deeper with Nested U-Structure for Salient Object Detectioncited by 412
  12. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 403
38 more
  1. 2021GMFlow: Learning Optical Flow via Global Matchingcited by 327
  2. 2023Effective Whole-body Pose Estimation with Two-stages Distillationcited by 311
  3. 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
  4. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 272
  5. 2023Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translationcited by 253
  6. 2019PointNetLK: Robust & Efficient Point Cloud Registration using PointNetcited by 233
  7. 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 224
  8. 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 200
  9. 2024MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidancecited by 198
  10. 2022MotionBERT: A Unified Perspective on Learning Human Motion Representationscited by 193
  11. 2020RPM-Net: Robust Point Matching using Learned Featurescited by 182
  12. 2017Flow-Guided Feature Aggregation for Video Object Detectioncited by 177
  13. 2023RTMPose: Real-Time Multi-Person Pose Estimation based on MMPosecited by 175
  14. 2020Gait Recognition via Effective Global-Local Feature Representation and Local Temporal Aggregationcited by 121
  15. 2021GaitGraph: Graph Convolutional Network for Skeleton-Based Gait Recognitioncited by 115
  16. 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimationcited by 114
  17. 2021Evidential Deep Learning for Open Set Action Recognitioncited by 96
  18. 2022OpenGait: Revisiting Gait Recognition Toward Better Practicalitycited by 85
  19. 2017Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognitioncited by 82
  20. 2022CRAFT: Cross-Attentional Flow Transformer for Robust Optical Flowcited by 82
  21. 2023VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimationcited by 82
  22. 2023AntGPT: Can Large Language Models Help Long-term Action Anticipation from Videos?cited by 81
  23. 2019Gait Recognition via Disentangled Representation Learningcited by 70
  24. 2021GaitSet: Cross-view Gait Recognition through Utilizing Gait as a Deep Setcited by 57
  25. 2022SKFlow: Learning Optical Flow with Super Kernelscited by 56
  26. 2023A Dynamic Multi-Scale Voxel Flow Network for Video Predictioncited by 53
  27. 2019HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimationcited by 50
  28. 2022Learning 3D Human Pose Estimation from Dozens of Datasets using a Geometry-Aware Autoencoder to Bridge Between Skeleton Formatscited by 50
  29. 2023Explicit Box Detection Unifies End-to-End Multi-Person Pose Estimationcited by 48
  30. 2024BigGait: Learning Gait Representation You Want by Large Vision Modelscited by 45
  31. 2023Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detectioncited by 43
  32. 2019Learning Temporal Pose Estimation from Sparsely-Labeled Videoscited by 39
  33. 2022Video Test-Time Adaptation for Action Recognitioncited by 35
  34. 2023Parsing is All You Need for Accurate Gait Recognition in the Wildcited by 32
  35. 2019Learning Multi-Human Optical Flowcited by 26
  36. 2023Video-FocalNets: Spatio-Temporal Focal Modulation for Video Action Recognitioncited by 18
  37. 2019View-invariant Deep Architecture for Human Action Recognition using late fusioncited by 15
  38. 2021Action2video: Generating Videos of Human 3D Actionscited by 15

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 3.

8 papers.

  1. 2025BiggerGait: Unlocking Gait Recognition with Layer-wise Representations from Large Vision Modelscited by 20
  2. 2025LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routingcited by 18
  3. 2025Procedural Mistake Detection via Action Effect Modelingcited by 6
  4. 2025Silhouette-based Gait Foundation Modelcited by 2
  5. 2026ViTs for Action Classification in Videos: An Approach to Risky Tackle Detection in American Football Practice Videoscited by 1
  6. 2025Context-Aware Network Based on Multi-scale Spatio-temporal Attention for Action Recognition in Videoscited by 0
  7. 2026H-Flow: Self-supervised Human Scene Flow via Physics-inspired Joint Multi-modal Learningcited by 0
  8. 2026Pose-Anchored Optical Flow for Low-Latency Human Action Anticipation in Human-Robot Teamingcited by 0