HuLiGen: 从参数化人体模型生成人类LiDAR点云
HuLiGen: Human LiDAR Generation from Parametric Body Models
- LIGM, CNRS, Univ Gustave Eiffel, ENPC, Institut Polytechnique de Paris（LIGM，法国国家科学研究中心，古斯塔夫·埃菲尔大学，巴黎高科路桥学校，巴黎综合理工学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对人类LiDAR点云稀缺问题，提出生成模型HuLiGen，从参数化人体模型生成点云，并用于合成预训练，在低数据下将MPJPE降低高达50%。
AI中文摘要：
人类LiDAR点云的采集和标注成本极高，因此成为一种稀缺资源，阻碍了基于该模态的人体分析的发展。为缓解这一稀缺性，先前的工作依赖于模拟的人类LiDAR数据，但这些样本未能完全反映真实观测的几何和传感特性。相比之下，我们提出了HuLiGen，一种生成模型，它利用基于流匹配目标训练的point transformer，从参数化人体模型生成人类LiDAR点云。我们证明，生成的点云更接近真实采集分布。利用HuLiGen生成合成数据，我们提出了一种仅使用合成数据的预训练方案，用于基于LiDAR的人体姿态估计（HPE），该方案达到了最先进的性能，在低标注和低数据场景下提升尤为显著，其中MPJPE降低了高达50%。代码、模型和生成的样本可在该https URL获取。
英文摘要：
LiDAR point clouds of humans are extremely expensive to collect and annotate, thus represent a scarce resource that hinders the development of human analysis using this modality. To alleviate this scarcity, prior work relies on simulated human LiDAR, but such samples do not fully reflect the geometry and sensing characteristics of real observations. In contrast, we introduce HuLiGen, a generative model that generates human LiDAR point clouds from a parametric body model, using a point transformer trained with a flow-matching objective. We show that our generated point clouds are closer to the real capture distribution. Using HuLiGen to generate synthetic data, we propose a synthetic-only pretraining scheme for LiDAR-based HPE that achieves state-of-the-art performance, with even larger gains in low-annotation and low-data regimes, where MPJPE is reduced by up to 50%. Code, models and generated samples are available at https://github.com/valeoai/HuLiGen.
