发表机构
Jiangsu University（江苏大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究基于视觉的低功耗边缘平台实时跌倒检测问题，提出基于物理知识的框架，引入双LTC架构，包含可学习耦合模块等，能连续模拟轨迹演变等，编码连续时间机械惯性，在两类数据集上验证核心能力，具竞争力准确性与物理可解释性。
AI 中文摘要
跌倒检测对老年人护理和智能监控至关重要，但现有基于视觉的方法主要将其视为静态姿势分类或离散时间模式匹配，忽略了人体支撑系统的不稳定动态。本文提出一个基于物理知识的跌倒检测框架，将跌倒重新定义为耦合动力系统中的稳定性丧失事件。引入了一种新颖的双LTC架构，包括质心（CoM）子系统和支撑面（BoS）子系统，均实例化为液体时间常数（LTC）神经网络，以通过自适应时间常数连续模拟惯性轨迹演变和地面接触调整。一个可学习的耦合模块模拟两个子系统之间的物理相互作用，而稳定性流形分类器在联合潜在空间中运行，通过受李雅普诺夫启发的稳定性指标检测边界穿越。互补的反事实轨迹投影和碰撞时间（TTC）估计进一步实现不可逆性评估和预警。该架构旨在支持三态预测范式（正常、跌倒、已跌倒）；在本初步研究中，我们在两类数据集（正常与跌倒）上验证了核心稳定性判别能力，完整的三态时间转换留待未来工作。与传统的CNN-RNN管道不同，所提出的公式编码了连续时间机械惯性，产生了一个参数少于50K的网络，能够在资源受限的边缘设备上进行实时推理。大量实验证明了其具有竞争力的准确性和卓越的物理可解释性，验证了其在低计算量视觉跌倒检测中的有效性。
英文摘要
Falling detection is vital for elderly care and intelligent surveillance; however, prevailing vision-based approaches predominantly frame it as static pose classification or discrete temporal pattern matching, fundamentally overlooking the instability dynamics of the human support system. This paper proposes a physics-informed falling detection framework that recasts falling as a stability-loss event in a coupled dynamical system. We introduce a novel dual-LTC architecture comprising a Center-of-Mass (CoM) subsystem and a Base-of-Support (BoS) subsystem, both instantiated as Liquid Time-Constant (LTC) neural networks to continuously model inertial trajectory evolution and ground-contact adjustment through adaptive time constants, Physical interpretability of falling motion. A learnable coupling module emulates physical interaction between the two subsystems, while a Stability Manifold classifier operates in the joint latent space to detect boundary crossing via Lyapunov-inspired stability metrics. Complementary counterfactual trajectory projection and Time-to-Collision (TTC) estimation further enable irreversibility assessment and early warning. The architecture is designed to support a three-state prediction paradigm (Normal, Falling, Fallen); in this preliminary study, we validate the core stability discrimination capability on a two-class dataset (Normal vs. Falling), leaving the full three-state temporal transition to future work. Unlike conventional CNN--RNN pipelines, the proposed formulation encodes continuous-time mechanical inertia, yielding a sub-50K-parameter network capable of real-time inference on resource-constrained edge devices. Extensive experiments demonstrate competitive accuracy with superior physical interpretability, validating its efficacy for low-compute visual fall detection.
