mlsanityData to 14 Oct 2026

3D-Aware Neural Body Fitting for Occlusion Robust 3D Human Pose Estimation

arXiv 2308.10123 ↗First posted 19 Aug 2023cs.CV cs.AIcited by 28

Not on the map. Why?

Abstract (arXiv)

Regression-based methods for 3D human pose estimation directly predict the 3D pose parameters from a 2D image using deep networks. While achieving state-of-the-art performance on standard benchmarks, their performance degrades under occlusion. In contrast, optimization-based methods fit a parametric body model to 2D features in an iterative manner. The localized reconstruction loss can potentially make them robust to occlusion, but they suffer from the 2D-3D ambiguity. Motivated by the recent success of generative models in rigid object pose estimation, we propose 3D-aware Neural Body Fitting (3DNBF) - an approximate analysis-by-synthesis approach to 3D human pose estimation with SOTA performance and occlusion robustness. In particular, we propose a generative model of deep features based on a volumetric human representation with Gaussian ellipsoidal kernels emitting 3D pose-dependent feature vectors. The neural features are trained with contrastive learning to become 3D-aware and hence to overcome the 2D-3D ambiguity. Experiments show that 3DNBF outperforms other approaches on both occluded and standard benchmarks. Code is available at https://github.com/edz-o/3DNBF

Key facts. Cited by 28 papers in the map (data through 14 Oct 2026). Builds on 63 of its 85 references that are in the map. First posted on arXiv 19 Aug 2023.

Builds on

63 of 85 references.

References · 63

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  5. 2013Describing Textures in the Wildcited by 2,125
  6. 2018Unsupervised Feature Learning via Non-Parametric Instance-level Discriminationcited by 1,845
  7. 2019Expressive Body Capture: 3D Hands, Face, and Body from a Single Imagecited by 1,612
  8. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,198
  9. 2016Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,064
  10. 2017End-to-end Recovery of Human Shape and Posecited by 885
  11. 2019PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitizationcited by 818
  12. 2020GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fieldscited by 785
38 more
  1. 2020Accelerating 3D Deep Learning with PyTorch3Dcited by 784
  2. 2016Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Imagecited by 695
  3. 2020Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humanscited by 634
  4. 2019Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loopcited by 627
  5. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
  6. 2022HumanNeRF: Free-viewpoint Rendering of Moving People from Monocular Videocited by 483
  7. 2019Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoningcited by 420
  8. 2021PARE: Part Attention Regressor for 3D Human Body Estimationcited by 337
  9. 2021ICON: Implicit Clothed humans Obtained from Normalscited by 283
  10. 2021Mesh Graphormercited by 265
  11. 2017Integral Human Pose Regressioncited by 256
  12. 2022NeuMan: Neural Human Radiance Field from a Single Videocited by 244
  13. 2018Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodiescited by 239
  14. 2021A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Posecited by 233
  15. 2018Learning to Estimate 3D Human Pose and Shape from a Single Color Imagecited by 226
  16. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
  17. 2017Unite the People: Closing the Loop Between 3D and 2D Human Representationscited by 205
  18. 2021Neural Articulated Radiance Fieldcited by 196
  19. 2017Single-Shot Multi-Person 3D Pose Estimation From Monocular RGBcited by 195
  20. 2018Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose and Shape Estimationcited by 185
  21. 2018Monocular Total Capture: Posing Face, Body, and Hands in the Wildcited by 162
  22. 2021Probabilistic Modeling for Human Mesh Recoverycited by 153
  23. 2017Towards 3D Human Pose Estimation in the Wild: a Weakly-supervised Approachcited by 115
  24. 2020Exemplar Fine-Tuning for 3D Human Model Fitting Towards In-the-Wild 3D Human Pose Estimationcited by 113
  25. 2019DenseRaC: Joint 3D Pose and Shape Estimation by Dense Render-and-Comparecited by 100
  26. 2017Self-supervised Learning of Motion Capturecited by 97
  27. 2020Weakly Supervised 3D Human Pose and Shape Reconstruction with Normalizing Flowscited by 94
  28. 2017Towards Accurate Markerless Human Shape and Pose Estimation over Timecited by 92
  29. 2020Human Body Model Fitting by Learned Gradient Descentcited by 81
  30. 2020Continuous Surface Embeddingscited by 80
  31. 2020Hierarchical Kinematic Human Mesh Recoverycited by 71
  32. 2019XNect: Real-time Multi-Person 3D Motion Capture with a Single RGB Cameracited by 68
  33. 2021RePOSE: Fast 6D Object Pose Refinement via Deep Texture Renderingcited by 63
  34. 20203D Human Pose Estimation using Spatio-Temporal Networks with Explicit Occlusion Trainingcited by 54
  35. 20203D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Datacited by 50
  36. 2021Hierarchical Kinematic Probability Distributions for 3D Human Shape and Pose Estimation from Images in the Wildcited by 48
  37. 2020Compositional Convolutional Neural Networks: A Deep Architecture with Innate Robustness to Partial Occlusioncited by 47
  38. 2016A Versatile Scene Model with Differentiable Visibility Applied to Generative Pose Estimationcited by 42

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 11; 2025: 10; 2026: 5.

28 papers.

Sort:

By area:

  1. 2024pix2gestalt: Amodal Segmentation by Synthesizing Wholescited by 70
  2. 2024RoHM: Robust Human Motion Reconstruction via Diffusioncited by 48
  3. 2024PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videoscited by 28
  4. 2025Learning Occlusion-Robust Vision Transformers for Real-Time UAV Trackingcited by 22
  5. 2025CoMotion: Concurrent Multi-person 3D Motioncited by 13
  6. 2024Application of Multimodal Fusion Deep Learning Model in Disease Recognitioncited by 12
  7. 2024Exploration of Attention Mechanism-Enhanced Deep Learning Models in the Mining of Medical Textual Datacited by 11
  8. 2024Enhancing Medical Imaging with GANs Synthesizing Realistic Images from Limited Datacited by 11
  9. 2024Research on the Application of Computer Vision Based on Deep Learning in Autonomous Driving Technologycited by 9
  10. 2023Wild2Avatar: Rendering Humans Behind Occlusionscited by 8
  11. 2024Integrating Medical Imaging and Clinical Reports Using Multimodal Deep Learning for Advanced Disease Analysiscited by 8
  12. 2025HiPART: Hierarchical Pose AutoRegressive Transformer for Occluded 3D Human Pose Estimationcited by 8
16 more
  1. 2025CHROME: Clothed Human Reconstruction with Occlusion-Resilience and Multiview-Consistency from a Single Imagecited by 6
  2. 2024Investigation of Customized Medical Decision Algorithms Utilizing Graph Neural Networkscited by 5
  3. 2024Multi-view Pose Fusion for Occlusion-Aware 3D Human Pose Estimationcited by 5
  4. 2025AJAHR: Amputated Joint Aware 3D Human Mesh Recoverycited by 3
  5. 2023STRIDE: Single-video based Temporally Continuous Occlusion-Robust 3D Pose Estimationcited by 2
  6. 2024Di2Pose: Discrete Diffusion Model for Occluded 3D Human Pose Estimationcited by 2
  7. 2025Application of Vision-Language Model to Pedestrians Behavior and Scene Understanding in Autonomous Drivingcited by 2
  8. 2026Masked Modeling for Human Motion Recovery Under Occlusionscited by 1
  9. 2026CIGPose: Causal Intervention Graph Neural Network for Whole-Body Pose Estimationcited by 1
  10. 2025First-place Solution for Streetscape Shop Sign Recognition Competitioncited by 0
  11. 2025DeProPose: Deficiency-Proof 3D Human Pose Estimation via Adaptive Multi-View Fusioncited by 0
  12. 2025DINeMo: Learning Neural Mesh Models with no 3D Annotationscited by 0
  13. 2025VOccl3D: A Video Benchmark Dataset for 3D Human Pose and Shape Estimation under real Occlusionscited by 0
  14. 2026OAHuman: Occlusion-Aware 3D Human Reconstruction from Monocular Imagescited by 0
  15. 2026Discriminative-Generative Synergy for Occlusion Robust 3D Human Mesh Recoverycited by 0
  16. 2026HumanFlow -- Diffusion-Driven MAV Navigation Among Humans via Tightly-Coupled Motion Tracking, Forecasting, and Controlcited by 0