mlsanityData to 7 Oct 2026

HiPART: Hierarchical Pose AutoRegressive Transformer for Occluded 3D Human Pose Estimation

arXiv 2503.23331 ↗First posted 30 Mar 2025cs.CV cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

Existing 2D-to-3D human pose estimation (HPE) methods struggle with the occlusion issue by enriching information like temporal and visual cues in the lifting stage. In this paper, we argue that these methods ignore the limitation of the sparse skeleton 2D input representation, which fundamentally restricts the 2D-to-3D lifting and worsens the occlusion issue. To address these, we propose a novel two-stage generative densification method, named Hierarchical Pose AutoRegressive Transformer (HiPART), to generate hierarchical 2D dense poses from the original sparse 2D pose. Specifically, we first develop a multi-scale skeleton tokenization module to quantize the highly dense 2D pose into hierarchical tokens and propose a Skeleton-aware Alignment to strengthen token connections. We then develop a Hierarchical AutoRegressive Modeling scheme for hierarchical 2D pose generation. With generated hierarchical poses as inputs for 2D-to-3D lifting, the proposed method shows strong robustness in occluded scenarios and achieves state-of-the-art performance on the single-frame-based 3D HPE. Moreover, it outperforms numerous multi-frame methods while reducing parameter and computational complexity and can also complement them to further enhance performance and robustness.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 48 of its 66 references that are in the map. First posted on arXiv 30 Mar 2025.

Builds on

48 of 66 references.

References · 48

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  4. 2017Neural Discrete Representation Learningcited by 4,650
  5. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  6. 2016Wide Residual Networkscited by 3,057
  7. 2013Describing Textures in the Wildcited by 2,110
  8. 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,085
  9. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,441
  10. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  11. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  12. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,155
36 more
  1. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,005
  2. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 633
  3. 2019Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loopcited by 625
  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. 2022DiffusionDet: Diffusion Model for Object Detectioncited by 354
  7. 2017Cascaded Pyramid Network for Multi-Person Pose Estimationcited by 341
  8. 2021PARE: Part Attention Regressor for 3D Human Body Estimationcited by 334
  9. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 271
  10. 2022Diffsound: Discrete Diffusion Model for Text-to-sound Generationcited by 256
  11. 2020Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Posecited by 248
  12. 2019Semantic Graph Convolutional Networks for 3D Human Pose Regressioncited by 204
  13. 2018Neural Body Fitting: Unifying Deep Learning and Model-Based Human Pose and Shape Estimationcited by 185
  14. 2022MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Videocited by 158
  15. 2021MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimationcited by 151
  16. 2022ProDiff: Progressive Fast Diffusion Model For High-Quality Text-to-Speechcited by 147
  17. 2021Generating Diverse Structure for Image Inpainting With Hierarchical VQ-VAEcited by 106
  18. 2022DiffPose: Toward More Reliable 3D Pose Estimationcited by 99
  19. 2019Unsupervised 3D Pose Estimation with Geometric Self-Supervisioncited by 95
  20. 2021Graph Stacked Hourglass Networks for 3D Human Pose Estimationcited by 94
  21. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 81
  22. 2022DiffPose: Multi-hypothesis Human Pose Estimation using Diffusion modelscited by 65
  23. 2025HaWoR: World-Space Hand Motion Reconstruction from Egocentric Videoscited by 62
  24. 2023Frequency-Aware Transformer for Learned Image Compressioncited by 56
  25. 2023Human Pose as Compositional Tokenscited by 45
  26. 2023DiffPose: SpatioTemporal Diffusion Model for Video-Based Human Pose Estimationcited by 36
  27. 2022DiffuPose: Monocular 3D Human Pose Estimation via Denoising Diffusion Probabilistic Modelcited by 29
  28. 20233D-Aware Neural Body Fitting for Occlusion Robust 3D Human Pose Estimationcited by 28
  29. 2019Feature Boosting Network For 3D Pose Estimationcited by 26
  30. 2020Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimationcited by 26
  31. 2020Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videoscited by 23
  32. 2023Pose-Oriented Transformer with Uncertainty-Guided Refinement for 2D-to-3D Human Pose Estimationcited by 19
  33. 2021Hierarchical Graph Networks for 3D Human Pose Estimationcited by 11
  34. 2023ActionPrompt: Action-Guided 3D Human Pose Estimation With Text and Pose Promptingcited by 8
  35. 2023Lifting by Image -- Leveraging Image Cues for Accurate 3D Human Pose Estimationcited by 6
  36. 2024Di2Pose: Discrete Diffusion Model for Occluded 3D Human Pose Estimationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 5.

8 papers.

  1. 2025HAODiff: Human-Aware One-Step Diffusion via Dual-Prompt Guidancecited by 5
  2. 2025A Structure-aware and Motion-adaptive Framework for 3D Human Pose Estimation with Mambacited by 3
  3. 2025FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimationcited by 1
  4. 2026CIGPose: Causal Intervention Graph Neural Network for Whole-Body Pose Estimationcited by 1
  5. 2026HYPERPOSE: Hyperbolic Kinematic Phase-Space Attention for 3D Human Pose Estimationcited by 0
  6. 2026H-Flow: Self-supervised Human Scene Flow via Physics-inspired Joint Multi-modal Learningcited by 0
  7. 2026MotionMAR: Multi-scale Auto-Regressive Human Motion Reconstruction from Sparse Observationscited by 0
  8. 2026A Plug-and-Play 2D Motion Interface for Real-World Motion Language Modelscited by 0