{
  "paper_id": "11033acdba99fd1a8ebd277802e5114d",
  "paper_title": "Pre-training a Density-Aware Pose Transformer for Robust LiDAR-based 3D Human Pose Estimation",
  "paper_authors": [
    "Xiaoqi An",
    "Lin Zhao",
    "Chen Gong",
    "Jun Li",
    "Jian Yang"
  ],
  "publication_date": "",
  "publication_venue": "unknown",
  "abstract": "With the rapid development of autonomous driving, LiDAR-based 3D Human Pose Estimation (3D HPE) is becoming a research focus. However, due to the noise and sparsity of LiDAR-captured point clouds, robust human pose estimation remains challenging. Most of the existing methods use temporal information, multi-modal fusion, or SMPL optimization to correct biased results. In this work, we try to obtain sufficient information for 3D HPE only by modeling the intrinsic properties of low-quality point clouds. Hence, a simple yet powerful method is proposed, which provides insights both on modeling and augmentation of point clouds. Specifically, we first propose a concise and effective density-aware pose transformer (DAPT) to get stable keypoint representations. By using a set of joint anchors and a carefully designed exchange module, valid information is extracted from point clouds with different densities. Then 1D heatmaps are utilized to represent the precise locations of the keypoints. Secondly, a comprehensive LiDAR human synthesis and augmentation method is proposed to pre-train the model, enabling it to acquire a better human body prior. We increase the diversity of point clouds by randomly sampling human positions and orientations and by simulating occlusions through the addition of laser-level masks. Extensive experiments have been conducted on multiple datasets, including IMU-annotated LidarHuman26M, SLOPER4D, and manually annotated Waymo Open Dataset v2.0 (Waymo), HumanM3. Our method demonstrates SOTA performance in all scenarios. In particular, compared with LPFormer on Waymo, we reduce the average MPJPE by 10.0mm. Compared with PRN on SLOPER4D, we notably reduce the average MPJPE by 20.7mm.",
  "keywords": [
    "LiDAR-based 3D human pose estimation",
    "Density-aware pose transformer",
    "Pre-training",
    "Point cloud",
    "Heatmap"
  ],
  "paper_institutions": [
    "PCA Lab, Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education, School of Computer Science and Engineering, Nanjing University of Science and Technology"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:46:13.020388Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本方法针对LiDAR点云噪声和稀疏性导致3D人体姿态估计不稳健的问题，提出了一种仅依赖单帧LiDAR点云的密度感知姿态变换器（DAPT）。DAPT通过可学习的关节锚点和多密度交换模块，从不同密度点云中稳定提取关键点特征，并使用XYZ轴上的1D热图精确定位关节。同时，设计了基于SMPL网格的光线投射合成与增广方法，通过随机场景采样、地面建模和激光级遮挡掩码生成多样化且贴近现实的预训练数据，使模型学习到更强的人体先验。在两阶段训练（预训练+微调）下，在LidarHuman26M、SLOPER4D、Waymo Open Dataset和HumanM3四个数据集中均取得了最先进性能，例如在Waymo上MPJPE比LPFormer降低10.0mm，在SLOPER4D上比PRN降低20.7mm。该方法无优化、无多模态或时序信息，具备实际部署潜力。主要创新包括联合锚点与MDE模块、热图解码以及综合预训练策略，为低质量点云下的姿态估计提供了有效解决方案。\n\n### 关键要点\n- 提出DAPT，通过可学习的关节锚点集在不同池化层级间交换信息，稳定提取关节表示。\n- 使用1D热图在XYZ轴上表示关节位置，比直接回归坐标更稳健，且解码时取峰值获得精确坐标。\n- 预训练阶段采用SMPL网格和光线投射生成合成点云，并随机采样人体位置和朝向以增加多样性。\n- 引入地面网格模拟真实场景，通过随机法向和大小生成，增强场景真实性。\n- 激光级掩码通过分割有效激光网格中的补丁并随机遮挡，模拟真实环境中的遮挡情况。\n- 预训练时采用坐标回归和点分割损失，微调时仅用热图KL散度损失，两阶段训练有效迁移人体先验。\n- 在Waymo数据集上MPJPE达到51.59mm，比LPFormer低10.0mm；在SLOPER4D上MPJPE达到28.01mm，比PRN低20.7mm。\n- 在HumanM3上MPJPE为59.76mm，比PRN低10.8mm；在LiDARHuman26M上PA-MPJPE比NE低3.2mm。\n- 稳定性评估表明，在点云抖动和噪声簇干扰下，DAPT性能下降更少，预训练进一步提升抗噪能力。\n- 消融实验显示，联合锚点和MDE模块使MPJPE降低16.2mm，热图解码再降低0.6mm；预训练整体带来7.5mm改善。\n- 与LIPD、FreeMotion、NoiseMotion等合成数据方法相比，本方法预训练效果最好。\n- 可视化显示，在遮挡、稀疏、多人物场景下，本方法均能稳定输出正确姿态，避免左右混淆。",
  "available_templates": [
    "deep_read",
    "simple"
  ]
}