面向低功耗边缘平台的实时视觉跌倒检测框架
Real-time fall detection based on vision for low-power edge platforms
AI 导读
Wenjun Xia 提出一种物理信息驱动的跌倒检测框架,将跌倒建模为耦合动力系统中的失稳事件,采用由重心(CoM)与支撑面(BoS)子系统构成的双 LTC 架构,配合稳定性流形分类器、反事实轨迹投影和 Time-to-Collision(TTC)估计实现早期预警。
正文 · 原文
This paper has been withdrawn by Wenjun Xia
No PDF available, click to view other formats
Abstract: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.
| Comments: | We have discovered a terminological inaccuracy in our manuscript. We are replacing the submission to correct this misnomer and revise the related context |
| Subjects: | Neurons and Cognition (q-bio.NC); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.12909 [q-bio.NC] |
| (or arXiv:2607.12909v2 [q-bio.NC] for this version) | |
| https://doi.org/10.48550/arXiv.2607.12909 arXiv-issued DOI via DataCite |
Submission history
From: Wenjun Xia [view email]
[v1]
Tue, 14 Jul 2026 15:43:41 UTC (337 KB)
[v2]
Tue, 17 Nov 2026 12:48:00 UTC (1 KB) (withdrawn)
来源:arXiv cs.AI · arxiv.org