AI Based Real Time Hand Gesture Recognition System
International Research Journal of Modernization in Engineering Technology and Science · 2024
This research presents a comprehensive approach for real-time hand gesture recognition using a synergistic combination of TensorFlow, OpenCV, and MediaPipe.Hand gesture recognition holds immense potential for natural and intuitive human-computer interaction in various applications, such as augmented reality, virtual reality, and humancomputer interfaces.The proposed system leverages the strengths of TensorFlow for deep learning-based model development, OpenCV for computer vision tasks, and MediaPipe for efficient hand landmark detection.The workflow begins with hand detection using OpenCV, followed by the extraction of hand landmarks through MediaPipe's hand tracking module.These landmarks serve as crucial input features for a customtrained TensorFlow model, designed to recognize a diverse set of hand gestures.The model is trained on a well-curated dataset, ensuring robust performance across different hand shapes, sizes, and orientations.