{
  "paper_id": "5ada0842d52f3e0312e2ee42e5d15a8f",
  "paper_title": "A Survey on 3D Egocentric Human Pose Estimation",
  "paper_authors": [
    "Md Mushfiqur Azam",
    "Kevin Desai"
  ],
  "publication_date": "",
  "publication_venue": "2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)",
  "abstract": "Egocentric human pose estimation aims to estimate human body poses and develop body representations from a first-person camera perspective. It has gained vast popularity in recent years because of its wide range of applications in sectors like XR-technologies, human-computer interaction, and fitness tracking. However, to the best of our knowledge, there is no systematic literature review based on the proposed solutions regarding egocentric 3D human pose estimation. To that end, the aim of this survey paper is to provide an extensive overview of the current state of egocentric pose estimation research. In this paper, we categorize and discuss the popular datasets and the different pose estimation models, highlighting the strengths and weaknesses of different methods by comparative analysis. This survey can be a valuable resource for both researchers and practitioners in the field, offering insights into key concepts and cutting-edge solutions in egocentric pose estimation, its wide-ranging applications, as well as the open problems with future scope.",
  "keywords": [
    "3D egocentric human pose estimation",
    "survey",
    "datasets",
    "skeletal methods",
    "body shape methods",
    "evaluation metrics"
  ],
  "paper_institutions": [
    "The University of Texas at San Antonio"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:46:14.302790Z",
  "template_tag": "simple",
  "summary": "### 摘要\n\n本综述针对3D自我中心人体姿态估计这一前沿领域进行了系统性梳理。该任务旨在从第一人称视角（如头戴式相机）估计人体姿态，在XR技术、人机交互、健身追踪等众多应用中具有重要价值。尽管已有针对传统2D/3D姿态估计的综述，但缺乏专门针对完整身体自我中心3D姿态估计的系统性文献回顾，本文填补了这一空白。文章首先介绍了9个公开数据集（如EgoCap、Mo^2Cap^2、xR-EgoPose、EgoBody等）的特点与用途，然后根据输出类型将约35种主流方法划分为基于骨架（skeletal）和基于人体形状（body shape）两大类，后者通常利用SMPL等参数模型。在评估方面，本文归纳了MPJPE、PA-MPJPE等常用指标，并在Mo^2Cap^2、xR-EgoPose和AMASS等数据集上进行了多方法性能对比。研究发现，基于2D-3D提升的骨架方法往往优于直接3D回归方法，而基于形状的方法在实时性和鲁棒性上各有优劣，如EgoPoser适合实时应用，AGRoL离线性能更佳。最后，文章指出了当前面临的挑战（如强遮挡、视野限制、数据稀缺）及未来方向，包括多视角一致性、时序上下文利用等，为研究者提供了全面的参考。\n\n### 关键要点\n\n- 自我中心姿态估计利用第一人称视角，在XR、人机交互和健康监测等领域具有广泛应用前景。\n- 目前缺乏针对完整身体自我中心3D姿态估计的系统综述，本文填补了这一研究空白。\n- 本文整理并描述了9个常用数据集，包括合成数据（如xR-EgoPose、EgoGTA）和真实数据（如EgoPW、Ego-Exo4D），各具不同规模、动作和场景特征。\n- 方法按输出形式分为基于骨架的方法（回归或热图）和基于人体形状的方法（如SMPL/SMPL-X参数模型）。\n- 基于骨架的方法中，2D-3D提升模型（如SelfPose-UNet）在Mo^2Cap^2上平均MPJPE达到46.61 mm，优于直接3D回归方法。\n- 基于形状的方法如AGRoL在AMASS数据集上实现了最低MPJPE（3.71 cm），但离线使用；EgoPoser则适合实时场景。\n- 自我中心姿态估计面临的主要挑战包括视角变化、深度信息缺失、自遮挡以及真实野外数据集稀缺。\n- 评估指标包括MPJPE、PA-MPJPE、BA-MPJPE、全局MPJPE、MPJRE、MPJVE、PCK以及头部平移和朝向误差等。\n- 性能对比显示，在xR-EgoPose数据集上Ego-STAN和EgoGlass等方法的MPJPE可达40 mm以下，显著优于早期方法。\n- 未来方向包括利用多视角一致性解决遮挡、利用时序和上下文信息增强模型鲁棒性，以及发展更具标准化的基准数据集。"
}