Smart Video Reproduction System Based on a Convolutional Neural Network (CCN)

Recent advances in computing and technological development have significantly enhanced the importance of algorithms such as neural networks, deep learning, and artificial intelligence in practical applications. Indeed, innovations like the Raspberry Pi and similar devices have facilitated neural network programming. This work details the development and implementation of a smart video reproduction system capable of recognizing various hand gestures, which are interpreted as instructions via a convolutional neural network (CNN). Using a presence sensor (PIR), the system determines when to activate. The CNN also enables realtime classification of content from a webcam, ensuring that the played videos align with user preferences. The integration of the PIR sensor and CNN presents an innovative approach for automated video playback management, enhancing user-system interaction and optimizing the viewing experience. The system efficiently responds to user presence and personalizes content based on real-time analysis. All functionalities are programmed on a Raspberry Pi model 4 computer via phyton software. The results demonstrate the effectiveness of the proposed approach, highlighting the potential of presence detection, content classification, and adaptive technologies based on computer vision.