🤖 AI Summary
To address high privacy risks in real-time elderly fall detection, the neglect of inter-frame joint dependencies in existing skeleton-based methods, model parameter redundancy, and evaluation bias under small-data regimes, this paper proposes a lightweight spatiotemporal graph convolutional network (ST-GCN). The method jointly models the temporal dynamics and spatial topological dependencies of skeletal joints, constructing a motion-dynamics-aware dynamic graph structure; model compression is achieved via channel pruning and hierarchical sparsity. Evaluated on three benchmark datasets—NTU RGB+D, NW-UCLA, and SYSU 3D—the approach achieves state-of-the-art performance: it reduces model parameters by 42%, improves average accuracy by over 3.2%, and eliminates raw video capture entirely, enabling image-free privacy-preserving fall recognition.
📝 Abstract
The increasing pace of population aging calls for better care and support systems. Falling is a frequent and critical problem for elderly people causing serious long-term health issues. Fall detection from video streams is not an attractive option for real-life applications due to privacy issues. Existing methods try to resolve this issue by using very low-resolution cameras or video encryption. However, privacy cannot be ensured completely with such approaches. Key points on the body, such as skeleton joints, can convey significant information about motion dynamics and successive posture changes which are crucial for fall detection. Skeleton joints have been explored for feature extraction but with image recognition models that ignore joint dependency across frames which is important for the classification of actions. Moreover, existing models are over-parameterized or evaluated on small datasets with very few activity classes. We propose an efficient graph convolution network model that exploits spatio-temporal joint dependencies and dynamics of human skeleton joints for accurate fall detection. Our method leverages dynamic representation with robust concurrent spatiotemporal characteristics of skeleton joints. We performed extensive experiments on three large-scale datasets. With a significantly smaller model size than most existing methods, our proposed method achieves state-of-the-art results on the large scale NTU datasets.