mlsanityData to 7 Oct 2026

2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning

arXiv 1802.09232 ↗First posted 26 Feb 2018cs.CVcited by 103

Not on the map. Why?

Abstract (arXiv)

Action recognition and human pose estimation are closely related but both problems are generally handled as distinct tasks in the literature. In this work, we propose a multitask framework for jointly 2D and 3D pose estimation from still images and human action recognition from video sequences. We show that a single architecture can be used to solve the two problems in an efficient way and still achieves state-of-the-art results. Additionally, we demonstrate that optimization from end-to-end leads to significantly higher accuracy than separated learning. The proposed architecture can be trained with data from different categories simultaneously in a seamlessly way. The reported results on four datasets (MPII, Human3.6M, Penn Action and NTU) demonstrate the effectiveness of our method on the targeted tasks.

Key facts. Cited by 103 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Feb 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 33; 2022: 22; 2023: 15; 2024: 17; 2025: 10; 2026: 6.

103 papers.

Sort:

  1. 2021Revisiting Skeleton-based Action Recognitioncited by 256
  2. 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 199
  3. 2022MotionBERT: A Unified Perspective on Learning Human Motion Representationscited by 188
  4. 2021Putting People in their Place: Monocular Regression of 3D People in Depthcited by 141
  5. 2021Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimationcited by 62
  6. 2025Visual Imitation Enables Contextual Humanoid Controlcited by 57
  7. 2021Semantics for Robotic Mapping, Perception and Interaction: A Surveycited by 56
  8. 2021Conditional Directed Graph Convolution for 3D Human Pose Estimationcited by 52
  9. 2021Recent Advances in Monocular 2D and 3D Human Pose Estimation: A Deep Learning Perspectivecited by 49
  10. 2021Pose Transformers (POTR): Human Motion Prediction with Non-Autoregressive Transformerscited by 46
  11. 2022STAR-Transformer: A Spatio-temporal Cross Attention Transformer for Human Action Recognitioncited by 43
  12. 2021When Human Pose Estimation Meets Robustness: Adversarial Algorithms and Benchmarkscited by 41
38 more
  1. 2021SportsCap: Monocular 3D Human Motion Capture and Fine-grained Understanding in Challenging Sports Videoscited by 40
  2. 2022Hierarchical Temporal Transformer for 3D Hand Pose Estimation and Action Recognition from Egocentric RGB Videoscited by 35
  3. 2022Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformerscited by 33
  4. 2021The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactionscited by 31
  5. 2021KAMA: 3D Keypoint Aware Body Mesh Articulationcited by 27
  6. 2022PedFormer: Pedestrian Behavior Prediction via Cross-Modal Attention Modulation and Gated Multitask Learningcited by 23
  7. 2022Cross-Modal Learning with 3D Deformable Attention for Action Recognitioncited by 20
  8. 2023Vision-Based Human Pose Estimation via Deep Learning: A Surveycited by 20
  9. 2021Fast Facial Landmark Detection and Applications: A Surveycited by 18
  10. 2021A Novel Dataset for Keypoint Detection of quadruped Animals from Imagescited by 18
  11. 2024Image anomaly detection and prediction scheme based on SSA optimized ResNet50-BiGRU modelcited by 18
  12. 2021Lifting Monocular Events to 3D Human Posescited by 16
  13. 2021HIH: Towards More Accurate Face Alignment via Heatmap in Heatmapcited by 15
  14. 2021Subpixel Heatmap Regression for Facial Landmark Localizationcited by 14
  15. 2023Reconstructing Three-Dimensional Models of Interacting Humanscited by 14
  16. 2024BaboonLand Dataset: Tracking Primates in the Wild and Automating Behaviour Recognition from Drone Videoscited by 14
  17. 2021Human Interaction Recognition Framework based on Interacting Body Part Attentioncited by 11
  18. 2021Localization with Sampling-Argmaxcited by 11
  19. 2021Unsupervised 3D Pose Estimation for Hierarchical Dance Video Recognitioncited by 9
  20. 20213D Pose Estimation and Future Motion Prediction from 2D Imagescited by 8
  21. 2022Interpretable Vertebral Fracture Quantification via Anchor-Free Landmarks Localizationcited by 8
  22. 2023ActionPrompt: Action-Guided 3D Human Pose Estimation With Text and Pose Promptingcited by 8
  23. 2023Ske2Grid: Skeleton-to-Grid Representation Learning for Action Recognitioncited by 7
  24. 2024From CNNs to Transformers in Multimodal Human Action Recognition: A Surveycited by 7
  25. 2021Enhanced 3D Human Pose Estimation from Videos by using Attention-Based Neural Network with Dilated Convolutionscited by 6
  26. 2022Skeleton Sequence and RGB Frame Based Multi-Modality Feature Fusion Network for Action Recognitioncited by 6
  27. 20223D Human Pose Estimation Using M"obius Graph Convolutional Networkscited by 6
  28. 2023CameraPose: Weakly-Supervised Monocular 3D Human Pose Estimation by Leveraging In-the-wild 2D Annotationscited by 6
  29. 2023Human Pose-based Estimation, Tracking and Action Recognition with Deep Learning: A Surveycited by 6
  30. 2024Deep Learning for Computer Vision based Activity Recognition and Fall Detection of the Elderly: a Systematic Reviewcited by 6
  31. 2024HumMUSS: Human Motion Understanding using State Space Modelscited by 6
  32. 2021Motion Projection Consistency Based 3D Human Pose Estimation with Virtual Bones from Monocular Videoscited by 5
  33. 2021Three-Stream 3D/1D CNN for Fine-Grained Action Classification and Segmentation in Table Tenniscited by 5
  34. 2021TEAM-Net: Multi-modal Learning for Video Action Recognition with Partial Decodingcited by 5
  35. 2022Video Violence Recognition and Localization Using a Semi-Supervised Hard Attention Modelcited by 5
  36. 2022CZU-MHAD: A multimodal dataset for human action recognition utilizing a depth camera and 10 wearable inertial sensorscited by 5
  37. 2023Human Body Pose Estimation for Gait Identification: A Comprehensive Survey of Datasets and Modelscited by 5
  38. 2021Win-Fail Action Recognitioncited by 4

50 most cited of 103 All 103 (JSON)