mlsanityData to 7 Oct 2026

ElePose: Unsupervised 3D Human Pose Estimation by Predicting Camera Elevation and Learning Normalizing Flows on 2D Poses

arXiv 2112.07088 ↗First posted 14 Dec 2021cs.CVcited by 35

Not on the map. Why?

Abstract (arXiv)

Human pose estimation from single images is a challenging problem that is typically solved by supervised learning. Unfortunately, labeled training data does not yet exist for many human activities since 3D annotation requires dedicated motion capture systems. Therefore, we propose an unsupervised approach that learns to predict a 3D human pose from a single image while only being trained with 2D pose data, which can be crowd-sourced and is already widely available. To this end, we estimate the 3D pose that is most likely over random projections, with the likelihood estimated using normalizing flows on 2D poses. While previous work requires strong priors on camera rotations in the training data set, we learn the distribution of camera angles which significantly improves the performance. Another part of our contribution is to stabilize training with normalizing flows on high-dimensional 3D pose data by first projecting the 2D poses to a linear subspace. We outperform the state-of-the-art unsupervised human pose estimation methods on the benchmark datasets Human3.6M and MPI-INF-3DHP in many metrics.

Key facts. Cited by 35 papers in the map (data through 7 Oct 2026). Builds on 48 of its 67 references that are in the map. First posted on arXiv 14 Dec 2021.

Builds on

48 of 67 references.

References · 48

  1. 2016Density estimation using Real NVPcited by 2,029
  2. 2019VIBE: Video Inference for Human Body Pose and Shape Estimationcited by 615
  3. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 493
  4. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  5. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 401
  6. 2017Cascaded Pyramid Network for Multi-Person Pose Estimationcited by 341
  7. 2021PARE: Part Attention Regressor for 3D Human Body Estimationcited by 334
  8. 2020HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimationcited by 304
  9. 2021Mesh Graphormercited by 264
  10. 2017Integral Human Pose Regressioncited by 255
  11. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 220
  12. 2021Probabilistic Modeling for Human Mesh Recoverycited by 152
36 more
  1. 2021SPEC: Seeing People in the Wild with an Estimated Cameracited by 134
  2. 2017VNect: Real-time 3D Human Pose Estimation with a Single RGB Cameracited by 129
  3. 2017Compositional Human Pose Regressioncited by 124
  4. 20163D Human Pose Estimation = 2D Pose Estimation + Matchingcited by 112
  5. 2018Ordinal Depth Supervision for 3D Human Pose Estimationcited by 107
  6. 2019Self-Supervised Learning of 3D Human Pose using Multi-view Geometrycited by 98
  7. 2019Unsupervised 3D Pose Estimation with Geometric Self-Supervisioncited by 95
  8. 2020Weakly Supervised 3D Human Pose and Shape Reconstruction with Normalizing Flowscited by 94
  9. 2017Exploiting temporal information for 3D pose estimationcited by 89
  10. 2019RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimationcited by 85
  11. 2021Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flowscited by 84
  12. 2020Cascaded deep monocular 3D human pose estimation with evolutionary training datacited by 80
  13. 20163D Human Pose Estimation from a Single Image via Distance Matrix Regressioncited by 73
  14. 2017Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Imagecited by 73
  15. 2018Learning Monocular 3D Human Pose Estimation from Multi-view Imagescited by 70
  16. 2019XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Cameracited by 68
  17. 2020Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wildcited by 66
  18. 2020CanonPose: Self-Supervised Monocular 3D Human Pose Estimation in the Wildcited by 65
  19. 2019HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimationcited by 50
  20. 2019In the Wild Human Pose Estimation Using Explicit 2D Features and Intermediate 3D Representationscited by 48
  21. 2019View-Invariant Probabilistic Embedding for Human Posecited by 47
  22. 2017A Perceptual Measure for Deep Single Image Camera Calibrationcited by 45
  23. 2018Can 3D Pose be Learned from 2D Projections Alone?cited by 44
  24. 2020Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesiscited by 44
  25. 2016Structured Prediction of 3D Human Pose with Deep Neural Networkscited by 43
  26. 2015Maximum-Margin Structured Learning with Deep Networks for 3D Human Pose Estimationcited by 37
  27. 2020Unsupervised 3D Human Pose Representation with Viewpoint and Pose Disentanglementcited by 32
  28. 2019Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimationcited by 31
  29. 20163D Human Pose Estimation Using Convolutional Neural Networks with 2D Pose Informationcited by 28
  30. 2018OriNet: A Fully Convolutional Network for 3D Human Pose Estimationcited by 28
  31. 2017Adversarial Inverse Graphics Networks: Learning 2D-to-3D Lifting and Image-to-Image Translation from Unpaired Supervisioncited by 27
  32. 2020Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimationcited by 26
  33. 2019Neural Scene Decomposition for Multi-Person Motion Capturecited by 21
  34. 2016Single Image 3D Interpreter Networkcited by 19
  35. 2019Distill Knowledge from NRSfM for Weakly Supervised 3D Pose Learningcited by 13
  36. 2019Weakly-Supervised 3D Pose Estimation from a Single Image using Multi-View Consistencycited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 14; 2024: 6; 2025: 8; 2026: 3.

35 papers.

Sort:

By area:

  1. 2022DiffPose: Multi-hypothesis Human Pose Estimation using Diffusion modelscited by 65
  2. 2023MAS: Multi-view Ancestral Sampling for 3D motion generation using 2D diffusioncited by 37
  3. 2023Back to Optimization: Diffusion-based Zero-Shot 3D Human Pose Estimationcited by 24
  4. 2022MoCapDeform: Monocular 3D Human Motion Capture in Deformable Scenescited by 18
  5. 2023Improving 3D Pose Estimation for Sign Languagecited by 16
  6. 2024Lifting Motion to the 3D World via 2D Diffusioncited by 11
  7. 2023Self-supervised 3D Human Pose Estimation from a Single Imagecited by 8
  8. 2023Unsupervised 3D Pose Estimation with Non-Rigid Structure-from-Motion Modelingcited by 7
  9. 2023LInKs "Lifting Independent Keypoints" -- Partial Pose Lifting for Occlusion Handling with Improved Accuracy in 2D-3D Human Pose Estimationcited by 6
  10. 2024StackFLOW: Monocular Human-Object Reconstruction by Stacked Normalizing Flow with Offsetcited by 5
  11. 2023Improving the Robustness of 3D Human Pose Estimation: A Benchmark and Learning from Noisy Inputcited by 4
  12. 2024PADS: Plug-and-Play 3D Human Pose Analysis via Diffusion Generative Modelingcited by 3
23 more
  1. 2022Unsupervised 3D Keypoint Discovery with Multi-View Geometrycited by 2
  2. 2023A Horse with no Labels: Self-Supervised Horse Pose Estimation from Unlabelled Images and Synthetic Priorcited by 2
  3. 2024STGFormer: Spatio-Temporal GraphFormer for 3D Human Pose Estimation in Videocited by 2
  4. 2025Every Angle Is Worth A Second Glance: Mining Kinematic Skeletal Structures from Multi-view Joint Cloudcited by 2
  5. 2025PersPose: 3D Human Pose Estimation with Perspective Encoding and Perspective Rotationcited by 2
  6. 2025Learning to Control Physically-simulated 3D Characters via Generating and Mimicking 2D Motionscited by 2
  7. 2026AnyLift: Scaling Motion Reconstruction from Internet Videos via 2D Diffusioncited by 2
  8. 2023Visualizing Skiers' Trajectories in Monocular Videoscited by 1
  9. 2023Unsupervised Multi-Person 3D Human Pose Estimation From 2D Poses Alonecited by 1
  10. 2024EPOCH: Jointly Estimating the 3D Pose of Cameras and Humanscited by 1
  11. 2025Spectral Compression Transformer with Line Pose Graph for Monocular 3D Human Pose Estimationcited by 1
  12. 2025Bimanual 3D Hand Motion and Articulation Forecasting in Everyday Imagescited by 1
  13. 2022Optimising 2D Pose Representation: Improve Accuracy, Stability and Generalisability Within Unsupervised 2D-3D Human Pose Estimationcited by 0
  14. 2023Learning from Abstract Images: on the Importance of Occlusion in a Minimalist Encoding of Human Posescited by 0
  15. 2023Mirror-Aware Neural Humanscited by 0
  16. 2023Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehiclecited by 0
  17. 2023Two Views Are Better than One: Monocular 3D Pose Estimation with Multiview Consistencycited by 0
  18. 2024Using Motion Cues to Supervise Single-Frame Body Pose and Shape Estimation in Low Data Regimescited by 0
  19. 2025Mocap-2-to-3: Multi-view Lifting for Monocular Motion Recovery with 2D Pretrainingcited by 0
  20. 2025Unsupervised Monocular 3D Keypoint Discovery from Multi-View Diffusion Priorscited by 0
  21. 2025LaxMotion: Rethinking Supervision Granularity for 3D Human Motion Generationcited by 0
  22. 2026Unsupervised 3D Human Pose Estimation via Conditional Multi-view Ancestral Samplingcited by 0
  23. 2026VideoMDM: Towards 3D Human Motion Generation From 2D Supervisioncited by 0