mlsanityData to 14 Oct 2026

Efficient Diffusion-Based 3D Human Pose Estimation with Hierarchical Temporal Pruning

arXiv 2508.21363 ↗First posted 29 Aug 2025cs.CVcited by 0

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

Abstract (arXiv)

Diffusion models have demonstrated strong capabilities in generating high-fidelity 3D human poses, yet their iterative nature and multi-hypothesis requirements incur substantial computational cost. In this paper, we propose an Efficient Diffusion-Based 3D Human Pose Estimation framework with a Hierarchical Temporal Pruning (HTP) strategy, which dynamically prunes redundant pose tokens across both frame and semantic levels while preserving critical motion dynamics. HTP operates in a staged, top-down manner: (1) Temporal Correlation-Enhanced Pruning (TCEP) identifies essential frames by analyzing inter-frame motion correlations through adaptive temporal graph construction; (2) Sparse-Focused Temporal MHSA (SFT MHSA) leverages the resulting frame-level sparsity to reduce attention computation, focusing on motion-relevant tokens; and (3) Mask-Guided Pose Token Pruner (MGPTP) performs fine-grained semantic pruning via clustering, retaining only the most informative pose tokens. Experiments on Human3.6M and MPI-INF-3DHP show that HTP reduces training MACs by 38.5%, inference MACs by 56.8%, and improves inference speed by an average of 81.1% compared to prior diffusion-based methods, while achieving state-of-the-art performance.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models. Builds on 40 of its 72 references that are in the map. First posted on arXiv 29 Aug 2025.

Builds on

40 of 72 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024FinePOSE: Fine-Grained Prompt-Driven 3D Human Pose Estimation via Diffusion Models8 May

Other references · 39

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
  4. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  5. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 915
  6. 2023MotionGPT: Human Motion as a Foreign Languagecited by 527
  7. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
  8. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 403
  9. 2017Cascaded Pyramid Network for Multi-Person Pose Estimationcited by 341
  10. 2021Come-Closer-Diffuse-Faster: Accelerating Conditional Diffusion Models for Inverse Problems through Stochastic Contractioncited by 307
  11. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 272
  12. 2022MoFusion: A Framework for Denoising-Diffusion-based Motion Synthesiscited by 218
27 more
  1. 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 200
  2. 2022MotionBERT: A Unified Perspective on Learning Human Motion Representationscited by 193
  3. 2022MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Videocited by 159
  4. 2021MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimationcited by 151
  5. 2020SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approachcited by 94
  6. 2023PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimationcited by 92
  7. 2019Multimodal Transformer with Multi-View Visual Representation for Image Captioningcited by 86
  8. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 82
  9. 2022P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimationcited by 80
  10. 2023MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Networkcited by 64
  11. 2020Pose-Guided Human Animation from a Single Image in the Wildcited by 53
  12. 2021Conditional Directed Graph Convolution for 3D Human Pose Estimationcited by 52
  13. 2023GLA-GCN: Global-local Adaptive Graph Convolutional Network for 3D Human Pose Estimation from Monocular Videocited by 47
  14. 2022Diffusion Video Autoencoders: Toward Temporally Consistent Face Video Editing via Disentangled Video Encodingcited by 39
  15. 2024KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimationcited by 34
  16. 20183D Human Pose Estimation in RGBD Images for Robotic Task Learningcited by 33
  17. 2022Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformerscited by 33
  18. 2022DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimationcited by 29
  19. 2023Co-Evolution of Pose and Mesh for 3D Human Body Estimation from Videocited by 25
  20. 2023HDFormer: High-order Directed Transformer for 3D Human Pose Estimationcited by 23
  21. 2024PoseMamba: Monocular 3D Human Pose Estimation with Bidirectional Global-Local Spatio-Temporal State Space Modelcited by 15
  22. 2025Heterogeneous Skeleton-Based Action Representation Learningcited by 12
  23. 2019View Invariant 3D Human Pose Estimationcited by 11
  24. 2025Foundation Model for Skeleton-Based Human Action Understandingcited by 11
  25. 2024Zero-Shot Skeleton-based Action Recognition with Dual Visual-Text Alignmentcited by 9
  26. 2025A Structure-aware and Motion-adaptive Framework for 3D Human Pose Estimation with Mambacited by 3
  27. 2024Graph and Skipped Transformer: Exploiting Spatial and Temporal Modeling Capacities for Efficient 3D Human Pose Estimationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation
  2. 2023DiffHPE: Robust, Coherent 3D Human Pose Lifting with Diffusion
  3. 2023Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation
  4. 2022A generic diffusion-based approach for 3D human pose prediction in the wild
  5. 2025H2OT: Hierarchical Hourglass Tokenizer for Efficient Video Pose Transformers
  6. 2025PRGCN: A Graph Memory Network for Cross-Sequence Pattern Reuse in 3D Human Pose Estimation
  7. 2023DiffPose: SpatioTemporal Diffusion Model for Video-Based Human Pose Estimation
  8. 2024Di2Pose: Discrete Diffusion Model for Occluded 3D Human Pose Estimation