LaDy：基于拉格朗日动力学先验的网络，通过时空调制实现骨架动作分割
LaDy: Lagrangian-Dynamic Informed Network for Skeleton-based Action Segmentation via Spatial-Temporal Modulation
把拉格朗日动力学中的广义力作为显式物理先验注入骨架序列建模，并用能量一致性损失约束学习，这一思路可迁移到其他需要区分「运动学相似但动力学意图不同」的时序密集预测任务（如细粒度动作识别、步态相位分割）。
原文依据 The learned dynamics then drive a Spatio-Temporal Modulation module
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骨架的时序动作分割（STAS）旨在将未裁剪的骨架序列密集解析为帧级动作类别。然而，现有方法虽擅长捕捉时空运动学特征，却忽略了支配人体运动的底层物理动力学。这一疏漏限制了对运动学相似但动力学意图不同的动作之间的类间判别力，也妨碍了在动力学力剖面发生转变处进行精确的边界定位。为解决这些问题，我们提出拉格朗日动力学先验网络（LaDy），一个将拉格朗日动力学原理融入分割过程的框架。具体而言，LaDy 首先由关节位置计算广义坐标，然后在物理约束下估计拉格朗日项，以显式合成广义力。为进一步保证物理一致性，我们的能量一致性损失强制满足功能原理，使动能变化与合力所做的功相一致。学习到的动力学随后驱动一个时空调制模块：在空间上，广义力与空间表征融合，以提供更具判别力的语义；在时序上，构建显著动力学信号用于时序门控，从而显著增强边界感知能力。在具有挑战性的数据集上的实验表明，LaDy 取得了最优性能，验证了将物理动力学融入动作分割的有效性。代码见 https://github.com/HaoyuJi/LaDy。
英文原文摘要 1535 字符
Skeleton-based Temporal Action Segmentation (STAS) aims to densely parse untrimmed skeletal sequences into frame-level action categories. However, existing methods, while proficient at capturing spatio-temporal kinematics, neglect the underlying physical dynamics that govern human motion. This oversight limits inter-class discriminability between actions with similar kinematics but distinct dynamic intents, and hinders precise boundary localization where dynamic force profiles shift. To address these, we propose the Lagrangian-Dynamic Informed Network (LaDy), a framework integrating principles of Lagrangian dynamics into the segmentation process. Specifically, LaDy first computes generalized coordinates from joint positions and then estimates Lagrangian terms under physical constraints to explicitly synthesize the generalized forces. To further ensure physical coherence, our Energy Consistency Loss enforces the work-energy theorem, aligning kinetic energy change with the work done by the net force. The learned dynamics then drive a Spatio-Temporal Modulation module: Spatially, generalized forces are fused with spatial representations to provide more discriminative semantics. Temporally, salient dynamic signals are constructed for temporal gating, thereby significantly enhancing boundary awareness. Experiments on challenging datasets show that LaDy achieves state-of-the-art performance, validating the integration of physical dynamics for action segmentation. Code is available at https://github.com/HaoyuJi/LaDy.