mlsanityData to 13 Oct 2026

Can 3D Pose be Learned from 2D Projections Alone?

arXiv 1808.07182 ↗First posted 22 Aug 2018cs.CVcited by 44

Not on the map. Why?

Abstract (arXiv)

3D pose estimation from a single image is a challenging task in computer vision. We present a weakly supervised approach to estimate 3D pose points, given only 2D pose landmarks. Our method does not require correspondences between 2D and 3D points to build explicit 3D priors. We utilize an adversarial framework to impose a prior on the 3D structure, learned solely from their random 2D projections. Given a set of 2D pose landmarks, the generator network hypothesizes their depths to obtain a 3D skeleton. We propose a novel Random Projection layer, which randomly projects the generated 3D skeleton and sends the resulting 2D pose to the discriminator. The discriminator improves by discriminating between the generated poses and pose samples from a real distribution of 2D poses. Training does not require correspondence between the 2D inputs to either the generator or the discriminator. We apply our approach to the task of 3D human pose estimation. Results on Human3.6M dataset demonstrates that our approach outperforms many previous supervised and weakly supervised approaches.

Key facts. Cited by 44 papers in the map (data through 13 Oct 2026). First posted on arXiv 22 Aug 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 15; 2022: 8; 2023: 10; 2024: 6; 2025: 3; 2026: 2.

44 papers.

Sort:

By area:

  1. 2021PoseAug: A Differentiable Pose Augmentation Framework for 3D Human Pose Estimationcited by 88
  2. 2023MAS: Multi-view Ancestral Sampling for 3D motion generation using 2D diffusioncited by 37
  3. 2021ElePose: Unsupervised 3D Human Pose Estimation by Predicting Camera Elevation and Learning Normalizing Flows on 2D Posescited by 35
  4. 2022Estimating Egocentric 3D Human Pose in the Wild with External Weak Supervisioncited by 33
  5. 2022Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformerscited by 33
  6. 2021AdaptPose: Cross-Dataset Adaptation for 3D Human Pose Estimation by Learnable Motion Generationcited by 32
  7. 2021KAMA: 3D Keypoint Aware Body Mesh Articulationcited by 27
  8. 2022PoseTriplet: Co-evolving 3D Human Pose Estimation, Imitation, and Hallucination under Self-supervisioncited by 25
  9. 2021Generalizable Human Pose Triangulationcited by 24
  10. 2024SelfPose3d: Self-Supervised Multi-Person Multi-View 3d Pose Estimationcited by 21
  11. 2023Global Adaptation meets Local Generalization: Unsupervised Domain Adaptation for 3D Human Pose Estimationcited by 19
  12. 2021MetaPose: Fast 3D Pose from Multiple Views without 3D Supervisioncited by 12
32 more
  1. 2024In My Perspective, In My Hands: Accurate Egocentric 2D Hand Pose and Action Recognitioncited by 12
  2. 2021Self-Supervised 3D Human Pose Estimation with Multiple-View Geometrycited by 10
  3. 2022On Triangulation as a Form of Self-Supervision for 3D Human Pose Estimationcited by 10
  4. 2021PAUL: Procrustean Autoencoder for Unsupervised Liftingcited by 9
  5. 2023Self-supervised 3D Human Pose Estimation from a Single Imagecited by 8
  6. 2023Weakly Supervised Multi-Modal 3D Human Body Pose Estimation for Autonomous Drivingcited by 8
  7. 2024Reconstructing Close Human Interactions from Multiple Viewscited by 7
  8. 2023CameraPose: Weakly-Supervised Monocular 3D Human Pose Estimation by Leveraging In-the-wild 2D Annotationscited by 6
  9. 2023LInKs "Lifting Independent Keypoints" -- Partial Pose Lifting for Occlusion Handling with Improved Accuracy in 2D-3D Human Pose Estimationcited by 6
  10. 2021DensePose 3D: Lifting Canonical Surface Maps of Articulated Objects to the Third Dimensioncited by 5
  11. 2021Learning Temporal 3D Human Pose Estimation with Pseudo-Labelscited by 4
  12. 2021Lifting 2D Human Pose to 3D with Domain Adapted 3D Body Conceptcited by 4
  13. 2023MPM: A Unified 2D-3D Human Pose Representation via Masked Pose Modelingcited by 4
  14. 2021SVMAC: Unsupervised 3D Human Pose Estimation from a Single Image with Single-view-multi-angle Consistencycited by 3
  15. 2021Self-supervised Secondary Landmark Detection via 3D Representation Learningcited by 3
  16. 2021TriPose: A Weakly-Supervised 3D Human Pose Estimation via Triangulation from Videocited by 2
  17. 2022Multi-person 3D pose estimation from unlabelled datacited by 2
  18. 2024Quater-GCN: Enhancing 3D Human Pose Estimation with Orientation and Semi-supervised Trainingcited by 2
  19. 2021Human Pose Manipulation and Novel View Synthesis using Differentiable Renderingcited by 1
  20. 2023Unsupervised Multi-Person 3D Human Pose Estimation From 2D Poses Alonecited by 1
  21. 2024EPOCH: Jointly Estimating the 3D Pose of Cameras and Humanscited by 1
  22. 2022"Teaching Independent Parts Separately" (TIPSy-GAN) : Improving Accuracy and Stability in Unsupervised Adversarial 2D to 3D Pose Estimationcited by 0
  23. 2022SPGNet: Spatial Projection Guided 3D Human Pose Estimation in Low Dimensional Spacecited by 0
  24. 2022Optimising 2D Pose Representation: Improve Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimationcited by 0
  25. 2023Novel View Synthesis of Humans using Differentiable Renderingcited by 0
  26. 2023Two Views Are Better than One: Monocular 3D Pose Estimation with Multiview Consistencycited by 0
  27. 2024Self-learning Canonical Space for Multi-view 3D Human Pose Estimationcited by 0
  28. 2025HDiffTG: A Lightweight Hybrid Diffusion-Transformer-GCN Architecture for 3D Human Pose Estimationcited by 0
  29. 2025Unsupervised Monocular 3D Keypoint Discovery from Multi-View Diffusion Priorscited by 0
  30. 2025LaxMotion: Rethinking Supervision Granularity for 3D Human Motion Generationcited by 0
  31. 2026Unsupervised 3D Human Pose Estimation via Conditional Multi-view Ancestral Samplingcited by 0
  32. 2026TransHands: Repurposing Human Pose Encoders as Hand Pose Encoderscited by 0