mlsanityData to 7 Oct 2026

On Triangulation as a Form of Self-Supervision for 3D Human Pose Estimation

arXiv 2203.15865 ↗First posted 29 Mar 2022cs.CVcited by 10

Not on the map. Why?

Abstract (arXiv)

Supervised approaches to 3D pose estimation from single images are remarkably effective when labeled data is abundant. However, as the acquisition of ground-truth 3D labels is labor intensive and time consuming, recent attention has shifted towards semi- and weakly-supervised learning. Generating an effective form of supervision with little annotations still poses major challenge in crowded scenes. In this paper we propose to impose multi-view geometrical constraints by means of a weighted differentiable triangulation and use it as a form of self-supervision when no labels are available. We therefore train a 2D pose estimator in such a way that its predictions correspond to the re-projection of the triangulated 3D pose and train an auxiliary network on them to produce the final 3D poses. We complement the triangulation with a weighting mechanism that alleviates the impact of noisy predictions caused by self-occlusion or occlusion from other subjects. We demonstrate the effectiveness of our semi-supervised approach on Human3.6M and MPI-INF-3DHP datasets, as well as on a new multi-view multi-person dataset that features occlusion.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 45 of its 65 references that are in the map. First posted on arXiv 29 Mar 2022.

Builds on

45 of 65 references.

References · 45

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2017Visualizing the Loss Landscape of Neural Netscited by 1,126
  5. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  6. 2017End-to-end Recovery of Human Shape and Posecited by 883
  7. 2017A Survey on Multi-Task Learningcited by 838
  8. 2019Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loopcited by 625
  9. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 493
  10. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  11. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 401
  12. 2016RMPE: Regional Multi-person Pose Estimationcited by 340
33 more
  1. 2018Learning 3D Human Dynamics from Videocited by 315
  2. 2017Integral Human Pose Regressioncited by 255
  3. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 220
  4. 2019Learnable Triangulation of Human Posecited by 160
  5. 2017VNect: Real-time 3D Human Pose Estimation with a Single RGB Cameracited by 129
  6. 2017Towards 3D Human Pose Estimation in the Wild: a Weakly-supervised Approachcited by 114
  7. 2018Hand Pose Estimation via Latent 2.5D Heatmap Regressioncited by 111
  8. 2018Ordinal Depth Supervision for 3D Human Pose Estimationcited by 107
  9. 2019Self-Supervised Learning of 3D Human Pose using Multi-view Geometrycited by 98
  10. 2019Unsupervised 3D Pose Estimation with Geometric Self-Supervisioncited by 95
  11. 2020SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approachcited by 94
  12. 2018LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Imagescited by 88
  13. 2021PoseAug: A Differentiable Pose Augmentation Framework for 3D Human Pose Estimationcited by 88
  14. 2019RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimationcited by 85
  15. 2020Cascaded deep monocular 3D human pose estimation with evolutionary training datacited by 80
  16. 2017Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Imagecited by 73
  17. 2018Unsupervised Geometry-Aware Representation for 3D Human Pose Estimationcited by 73
  18. 2018Learning Monocular 3D Human Pose Estimation from Multi-view Imagescited by 70
  19. 2020Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wildcited by 66
  20. 2020Lightweight Multi-View 3D Pose Estimation through Camera-Disentangled Representationcited by 57
  21. 2019TexturePose: Supervising Human Mesh Estimation with Texture Consistencycited by 53
  22. 2017Harvesting Multiple Views for Marker-less 3D Human Pose Annotationscited by 48
  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. 2017Adversarial PoseNet: A Structure-aware Convolutional Network for Human Pose Estimationcited by 35
  28. 2021TransFusion: Cross-view Fusion with Transformer for 3D Human Pose Estimationcited by 32
  29. 2019Multiview-Consistent Semi-Supervised Learning for 3D Human Pose Estimationcited by 31
  30. 2017Adversarial Inverse Graphics Networks: Learning 2D-to-3D Lifting and Image-to-Image Translation from Unpaired Supervisioncited by 27
  31. 2017Deep Multitask Architecture for Integrated 2D and 3D Human Sensingcited by 21
  32. 2019Distill Knowledge from NRSfM for Weakly Supervised 3D Pose Learningcited by 13
  33. 2014FollowMe: Efficient Online Min-Cost Flow Tracking with Bounded Memory and Computationcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 2; 2025: 0; 2026: 4.

10 papers.

  1. 2024MEGA: Masked Generative Autoencoder for Human Mesh Recoverycited by 13
  2. 2023Improved Trajectory Reconstruction for Markerless Pose Estimationcited by 12
  3. 2023Markerless Motion Capture and Biomechanical Analysis Pipelinecited by 10
  4. 2026Markerless Motion Capture for Biomechanical Whole-Body Kinematic Estimation in Infantscited by 1
  5. 2023Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehiclecited by 0
  6. 2023Two Views Are Better than One: Monocular 3D Pose Estimation with Multiview Consistencycited by 0
  7. 2024Occlusion Resilient 3D Human Pose Estimationcited by 0
  8. 2026Mesh-Aware Epipolar Matching for Multi-View Multi-Person 3D Pose Estimation in Basketballcited by 0
  9. 2026Biomechanics-aware Multi-view Markerless Motion Capture of Dexterous Hand Movementscited by 0
  10. 2026Cross-Model Distillation of a Human-Pose Foundation Model from Unannotated Infant Video for Markerless 3D Pose Estimationcited by 0