FastDDHPose: Towards Unified, Efficient, and Disentangled 3D Human Pose Estimation

2025-12-16

AI Summary

Existing monocular 3D human pose estimation methods lack a unified training and evaluation framework, hindering fair comparison and impeding training efficiency. To address this, we propose Fast3DHPE—the first standardized, fully reproducible end-to-end framework for monocular 3D human pose estimation. Building upon it, we introduce FastDDHPose, the first diffusion-based approach that explicitly decouples skeletal length and orientation distributions in the generative modeling process. It incorporates a kinematics-aware hierarchical spatiotemporal denoiser, jointly optimizing biomechanical plausibility and computational efficiency via hierarchical spatiotemporal attention and skeletal kinematic priors. Our method achieves state-of-the-art accuracy on Human3.6M and MPI-INF-3DHP while significantly accelerating training. Moreover, it demonstrates strong generalization and robustness in challenging in-the-wild scenarios. The code and pretrained models are publicly available.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Recent approaches for monocular 3D human pose estimation (3D HPE) have achieved leading performance by directly regressing 3D poses from 2D keypoint sequences. Despite the rapid progress in 3D HPE, existing methods are typically trained and evaluated under disparate frameworks, lacking a unified framework for fair comparison. To address these limitations, we propose Fast3DHPE, a modular framework that facilitates rapid reproduction and flexible development of new methods. By standardizing training and evaluation protocols, Fast3DHPE enables fair comparison across 3D human pose estimation methods while significantly improving training efficiency. Within this framework, we introduce FastDDHPose, a Disentangled Diffusion-based 3D Human Pose Estimation method which leverages the strong latent distribution modeling capability of diffusion models to explicitly model the distributions of bone length and bone direction while avoiding further amplification of hierarchical error accumulation. Moreover, we design an efficient Kinematic-Hierarchical Spatial and Temporal Denoiser that encourages the model to focus on kinematic joint hierarchies while avoiding unnecessary modeling of overly complex joint topologies. Extensive experiments on Human3.6M and MPI-INF-3DHP show that the Fast3DHPE framework enables fair comparison of all methods while significantly improving training efficiency. Within this unified framework, FastDDHPose achieves state-of-the-art performance with strong generalization and robustness in in-the-wild scenarios. The framework and models will be released at: https://github.com/Andyen512/Fast3DHPE

Problem

• Proposes a unified framework for fair comparison in 3D human pose estimation
• Introduces a disentangled diffusion method to model bone length and direction distributions
• Designs an efficient denoiser focusing on kinematic hierarchies to reduce error accumulation

Innovation

• Modular framework enabling fair comparison and efficient training
• Disentangled diffusion model separates bone length and direction distributions
• Kinematic-hierarchical denoiser focuses on joint hierarchies avoiding complex topologies