面向激光雷达人体运动捕捉的贝塞尔退化建模
Bezier Degradation Modeling for LiDAR-based Human Motion Capture
把人体运动序列用时间上可压缩的贝塞尔曲线表示，并用多尺度 Time-scale Motion Transformer 预测控制点，为点云输入下的遮挡补全与抖动抑制提供了可复用的轨迹参数化思路。
原文依据 we propose BMLiCap, a coarse-to-fine framework that models motion using temporally compressible Bezier curves
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论基于激光雷达的三维人体运动捕捉在自动驾驶、机器人等领域具有广泛应用，准确的运动重建在这些场景中至关重要。然而，现有方法往往难以应对不稳定的输入和严重遮挡，导致姿态预测出现抖动甚至失败。为应对这些挑战，我们提出 BMLiCap，一个由粗到细的框架，利用时间上可压缩的贝塞尔曲线对运动进行建模。通过一种保持轨迹的策略减少控制点，我们获得了连贯且易于学习的运动表示。为从激光雷达点云线索中重建人体动作，我们设计了一个渐进式运动重建模块。具体而言，引入时间尺度运动 Transformer（TMT）在多个时间尺度上预测运动曲线，并利用多级运动聚合器（MMA）自适应地融合多尺度曲线，以恢复细节丰富、时间上连贯的姿态，从而有效弥合由遮挡和噪声造成的观测空缺。在 LiDARHuman26M、FreeMotion、NoiseMotion 和 SLOPER4D 四个主流基准上，BMLiCap 在复杂场景中取得了最优的精度和时序连续性，展现出其补偿严重遮挡并减少预测抖动能力。
英文原文摘要 1282 字符
LiDAR-based 3D human motion capture has broad applications in fields such as autonomous driving and robotics, where accurate motion reconstruction is crucial. However, existing methods often struggle with unstable inputs and severe occlusions, leading to jittery or even failed pose predictions. To address these challenges, we propose BMLiCap, a coarse-to-fine framework that models motion using temporally compressible Bezier curves. By reducing control points through a trajectory-preserving strategy, we obtain a coherent and learning-friendly motion representation. To reconstruct human actions from LiDAR point-cloud cues, we design a progressive motion-reconstruction module. Specifically, a Time-scale Motion Transformer (TMT) is introduced to predict motion curves at multiple temporal scales, and a Multi-level Motion Aggregator (MMA) is utilized to adaptively fuse the multi-scale curves to recover detailed, temporally coherent poses, effectively bridging observation gaps caused by occlusions and noise. Across four mainstream benchmarks LiDARHuman26M, FreeMotion, NoiseMotion, and SLOPER4D, BMLiCap achieves state-of-the-art accuracy and temporal continuity in complex scenes, demonstrating its ability to compensate for severe occlusions and reduce prediction jitter.