Detecting driver fatigue is critical for road safety, as drowsy driving remains a leading cause of traffic accidents. Many existing solutions rely on computationally demanding deep learning models, which result in high latency and are unsuitable for embedded robotic devices with limited resources (such as intelligent vehicles/cars) where rapid detection is necessary to prevent accidents. This paper introduces LiteFat, a lightweight spatio-temporal graph learning model designed to detect driver fatigue efficiently while maintaining high accuracy and low computational demands. LiteFat involves converting streaming video data into spatio-temporal graphs (STG) using facial landmark detection, which focuses on key motion patterns and reduces unnecessary data processing. LiteFat uses MobileNet to extract facial features and create a feature matrix for the STG. A lightweight spatio-temporal graph neural network is then employed to identify signs of fatigue with minimal processing and low latency. Experimental results on benchmark datasets show that LiteFat performs competitively while significantly decreasing computational complexity and latency as compared to current state-of-the-art methods. This work enables the development of real-time, resource-efficient human fatigue detection systems that can be implemented upon embedded robotic devices.

The main contributions of this paper are as follows:

We propose a lightweight graph learning human fatigue detection model called LiteFat, which can achieve high accuracy while maintaining low computational complexity, making it suitable for real-time deployment on embedded robotic devices. % This paper deals with human fatigue detection in resource-constrained environments.
We introduce a multimodal fusion approach that dynamically constructs feature matrices from facial landmarks and MobileNet-based facial embeddings. Instead of using predefined facial graph structures, LiteFat learns the adjacency matrix adaptively during training, reducing human bias and capturing more accurate spatial relationships.
We conduct comprehensive experiments on a benchmark driver fatigue dataset. LiteFat demonstrates state-of-the-art (SOTA) performance, while significantly improving computational efficiency and reducing latency compared to baseline methods.