Key points are not available for this paper at this time.
The article introduces an approach for real-time sign language detection tailored for home automation applications. The proposed model utilizes the YOLOv8 (You Only Look Once) architecture for efficient and accurate sign detection while incorporating face recognition for identity verification. The fusion of sign language detection and identity verification enhances the system's usability in secure and personalized home automation environments. To address the challenge of deploying this system on low-resource hardware such as the Raspberry Pi, a unique approach is taken. The model leverages a combination of low-resolution and high-resolution images, optimizing the trade-off between accuracy and computational efficiency. This design choice ensures the feasibility of real-time sign language detection on resource-constrained devices commonly found in home automation setups. Furthermore, the system implements a frame skipping technique to reduce the frame rate, maintaining real-time responsiveness while further alleviating computational demands. This feature enhances the adaptability of the proposed model to diverse hardware configurations, making it suitable for a broad range of applications within the context of home automation. Impressive and consistent results were obtained at both experimental and implementation stages. It shows that this systems precision accuracy goes up to 97%.
Hussein et al. (Mon,) studied this question.