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

• URL: http://arxiv.org/abs/2508.21363v1
• Date: Fri, 29 Aug 2025 07:08:07 GMT
• Title: Efficient Diffusion-Based 3D Human Pose Estimation with Hierarchical Temporal Pruning
• Authors: Yuquan Bi, Hongsong Wang, Xinli Shi, Zhipeng Gui, Jie Gui, Yuan Yan Tang,
• Abstract summary: We propose an Efficient Diffusion-Based 3D Human Pose Estimation framework with a Temporal Pruning (HTP) strategy.<n>HTP prunes redundant pose tokens across both frame and semantic levels while preserving critical motion dynamics.<n>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.
• Score: 34.116532190562815
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation [32.94049816382114]
  We propose Fast3DHPE, a modular framework that facilitates rapid reproduction and flexible development of new methods.<n>By standardizing training and evaluation protocols, Fast3DHPE enables fair comparison across 3D human pose estimation methods.<n>Within this framework, we introduce FastDDHPose, a Disentangled Diffusion-based 3D Human Pose Estimation method.
  arXiv  Detail & Related papers  (2025-12-16T07:47:06Z)
• StarPose: 3D Human Pose Estimation via Spatial-Temporal Autoregressive Diffusion [29.682018018059043]
  StarPose is an autoregressive diffusion framework for 3D human pose estimation.<n>It incorporates historical 3D pose predictions and spatial-temporal physical guidance.<n>It achieves superior accuracy and temporal consistency in 3D human pose estimation.
  arXiv  Detail & Related papers  (2025-08-04T04:50:05Z)
• Memory-efficient Low-latency Remote Photoplethysmography through Temporal-Spatial State Space Duality [15.714133129768323]
  ME-r is a memory-efficient algorithm built on temporal-spatial state space duality.<n>It efficiently captures subtle periodic variations across facial frames while maintaining minimal computational overhead.<n>Our solution enables real-time inference with only 3.6 MB memory usage and 9.46 ms latency.
  arXiv  Detail & Related papers  (2025-04-02T14:34:04Z)
• ALOcc: Adaptive Lifting-based 3D Semantic Occupancy and Cost Volume-based Flow Prediction [89.89610257714006]
  Existing methods prioritize higher accuracy to cater to the demands of these tasks. We introduce a series of targeted improvements for 3D semantic occupancy prediction and flow estimation. Our purelytemporalal architecture framework, named ALOcc, achieves an optimal tradeoff between speed and accuracy.
  arXiv  Detail & Related papers  (2024-11-12T11:32:56Z)
• STGFormer: Spatio-Temporal GraphFormer for 3D Human Pose Estimation in Video [7.345621536750547]
  This paper presents the S-Temporal GraphFormer framework (STGFormer) for 3D human pose estimation in videos.<n>First, we introduce a STG attention mechanism, designed to more effectively leverage the inherent graph distributions of human body.<n>Next, we present a Modulated Hop-wise Regular GCN to independently process temporal and spatial dimensions in parallel.<n>Finally, we demonstrate our method state-of-the-art performance on the Human3.6M and MPIINF-3DHP datasets.
  arXiv  Detail & Related papers  (2024-07-14T06:45:27Z)
• LMD: Faster Image Reconstruction with Latent Masking Diffusion [28.54828478259779]
  Masked autoencoders (MAEs), as popular self-supervised vision learners, have demonstrated simpler and more effective image reconstruction and transfer capabilities on downstream tasks. This paper presents LMD, a faster image reconstruction framework with latent masking diffusion.
  arXiv  Detail & Related papers  (2023-12-13T08:36:51Z)
• Coordinate Transformer: Achieving Single-stage Multi-person Mesh Recovery from Videos [91.44553585470688]
  Multi-person 3D mesh recovery from videos is a critical first step towards automatic perception of group behavior in virtual reality, physical therapy and beyond. We propose the Coordinate transFormer (CoordFormer) that directly models multi-person spatial-temporal relations and simultaneously performs multi-mesh recovery in an end-to-end manner. Experiments on the 3DPW dataset demonstrate that CoordFormer significantly improves the state-of-the-art, outperforming the previously best results by 4.2%, 8.8% and 4.7% according to the MPJPE, PAMPJPE, and PVE metrics, respectively.
  arXiv  Detail & Related papers  (2023-08-20T18:23:07Z)
• Masked Motion Predictors are Strong 3D Action Representation Learners [143.9677635274393]
  In 3D human action recognition, limited supervised data makes it challenging to fully tap into the modeling potential of powerful networks such as transformers. We show that instead of following the prevalent pretext to perform masked self-component reconstruction in human joints, explicit contextual motion modeling is key to the success of learning effective feature representation for 3D action recognition.
  arXiv  Detail & Related papers  (2023-08-14T11:56:39Z)
• Back to MLP: A Simple Baseline for Human Motion Prediction [59.18776744541904]
  This paper tackles the problem of human motion prediction, consisting in forecasting future body poses from historically observed sequences. We show that the performance of these approaches can be surpassed by a light-weight and purely architectural architecture with only 0.14M parameters. An exhaustive evaluation on Human3.6M, AMASS and 3DPW datasets shows that our method, which we dub siMLPe, consistently outperforms all other approaches.
  arXiv  Detail & Related papers  (2022-07-04T16:35:58Z)
• P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation [78.83305967085413]
  This paper introduces a novel Pre-trained Spatial Temporal Many-to-One (P-STMO) model for 2D-to-3D human pose estimation task. Our method outperforms state-of-the-art methods with fewer parameters and less computational overhead.
  arXiv  Detail & Related papers  (2022-03-15T04:00:59Z)
• Consistency Guided Scene Flow Estimation [159.24395181068218]
  CGSF is a self-supervised framework for the joint reconstruction of 3D scene structure and motion from stereo video. We show that the proposed model can reliably predict disparity and scene flow in challenging imagery. It achieves better generalization than the state-of-the-art, and adapts quickly and robustly to unseen domains.
  arXiv  Detail & Related papers  (2020-06-19T17:28:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.