mlsanityData to 14 Oct 2026

Di2Pose: Discrete Diffusion Model for Occluded 3D Human Pose Estimation

arXiv 2405.17016 ↗First posted 27 May 2024cs.CVcited by 2

On the map: Diffusion models → 3D, point cloud and scene generation

Abstract (arXiv)

Continuous diffusion models have demonstrated their effectiveness in addressing the inherent uncertainty and indeterminacy in monocular 3D human pose estimation (HPE). Despite their strengths, the need for large search spaces and the corresponding demand for substantial training data make these models prone to generating biomechanically unrealistic poses. This challenge is particularly noticeable in occlusion scenarios, where the complexity of inferring 3D structures from 2D images intensifies. In response to these limitations, we introduce the Discrete Diffusion Pose (Di2Pose), a novel framework designed for occluded 3D HPE that capitalizes on the benefits of a discrete diffusion model. Specifically, Di2Pose employs a two-stage process: it first converts 3D poses into a discrete representation through a pose quantization step, which is subsequently modeled in latent space through a discrete diffusion process. This methodological innovation restrictively confines the search space towards physically viable configurations and enhances the model's capability to comprehend how occlusions affect human pose within the latent space. Extensive evaluations conducted on various benchmarks (e.g., Human3.6M, 3DPW, and 3DPW-Occ) have demonstrated its effectiveness.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models. Builds on 71 of its 92 references that are in the map. First posted on arXiv 27 May 2024.

Builds on

71 of 92 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023DiffHPE: Robust, Coherent 3D Human Pose Lifting with Diffusion4 Sep
  2. 2023DiffPose: SpatioTemporal Diffusion Model for Video-Based Human Pose Estimation31 Jul
  3. 2022DiffuPose: Monocular 3D Human Pose Estimation via Denoising Diffusion Probabilistic Model6 Dec
  4. 2022DiffPose: Multi-hypothesis Human Pose Estimation using Diffusion models29 Nov

Other references · 67

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  5. 2016Layer Normalizationcited by 5,669
  6. 2017Neural Discrete Representation Learningcited by 4,712
  7. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  8. 2013Describing Textures in the Wildcited by 2,125
  9. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  10. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,446
  11. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  12. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,163
38 more
  1. 2021Blended Diffusion for Text-driven Editing of Natural Imagescited by 927
  2. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 774
  3. 2022Autoregressive Image Generation using Residual Quantizationcited by 722
  4. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  5. 2022ViTPose: Simple Vision Transformer Baselines for Human Pose Estimationcited by 456
  6. 2023Finite Scalar Quantization: VQ-VAE Made Simplecited by 440
  7. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 403
  8. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  9. 2022DiffusionDet: Diffusion Model for Object Detectioncited by 354
  10. 2021PARE: Part Attention Regressor for 3D Human Body Estimationcited by 337
  11. 2020HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimationcited by 307
  12. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 272
  13. 2022Diffsound: Discrete Diffusion Model for Text-to-sound Generationcited by 257
  14. 2017Integral Human Pose Regressioncited by 256
  15. 2016Coarse-to-Fine Volumetric Prediction for Single-Image 3D Human Posecited by 221
  16. 2019Semantic Graph Convolutional Networks for 3D Human Pose Regressioncited by 204
  17. 2022ProDiff: Progressive Fast Diffusion Model For High-Quality Text-to-Speechcited by 149
  18. 2023LayoutDM: Discrete Diffusion Model for Controllable Layout Generationcited by 143
  19. 2021Human Pose Regression with Residual Log-likelihood Estimationcited by 131
  20. 2020Exemplar Fine-Tuning for 3D Human Model Fitting Towards In-the-Wild 3D Human Pose Estimationcited by 113
  21. 2021Generating Diverse Structure for Image Inpainting With Hierarchical VQ-VAEcited by 106
  22. 2021imGHUM: Implicit Generative Models of 3D Human Shape and Articulated Posecited by 103
  23. 20233D Human Pose Estimation via Intuitive Physicscited by 103
  24. 2022DiffPose: Toward More Reliable 3D Pose Estimationcited by 100
  25. 2021Graph Stacked Hourglass Networks for 3D Human Pose Estimationcited by 95
  26. 2023PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimationcited by 92
  27. 2017Exploiting temporal information for 3D pose estimationcited by 89
  28. 2020Motion Guided 3D Pose Estimation from Videoscited by 85
  29. 2023DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuningcited by 84
  30. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 82
  31. 2022P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimationcited by 80
  32. 2023Priority-Centric Human Motion Generation in Discrete Latent Spacecited by 67
  33. 2023Make-An-Animation: Large-Scale Text-conditional 3D Human Motion Generationcited by 56
  34. 2023Online Clustered Codebookcited by 55
  35. 2021Conditional Directed Graph Convolution for 3D Human Pose Estimationcited by 52
  36. 20203D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Datacited by 50
  37. 2023Human Pose as Compositional Tokenscited by 45
  38. 2022Poseur: Direct Human Pose Regression with Transformerscited by 43

50 most cited of 67 All 67 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025HiPART: Hierarchical Pose AutoRegressive Transformer for Occluded 3D Human Pose Estimationcited by 8
  2. 2026Flow6D: Discrete-to-Continuous Flow Matching for Efficient and Accurate Category-Level 6D Pose Estimationcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026Flexible Geometric Guidance for Probabilistic Human Pose Estimation with Diffusion Models
  2. 2025FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation
  3. 2022A generic diffusion-based approach for 3D human pose prediction in the wild
  4. 2024PADS: Plug-and-Play 3D Human Pose Analysis via Diffusion Generative Modeling
  5. 2024Disentangled Diffusion-Based 3D Human Pose Estimation with Hierarchical Spatial and Temporal Denoiser
  6. 2024D3PRefiner: A Diffusion-based Denoise Method for 3D Human Pose Refinement
  7. 2025DPoser-X: Diffusion Model as Robust 3D Whole-body Human Pose Prior
  8. 2024Exploring Latent Cross-Channel Embedding for Accurate 3D Human Pose Reconstruction in a Diffusion Framework