ARCHIVES

This paper presents the design and implementation of a gesture-controlled interface aimed at enhancing human- computer interaction (HCI) through intuitive and contactless control mechanisms. By leveraging computer vision and deep learning techniques, the proposed system recognizes hand gestures in real-time and maps them to specific system commands, enabling users to interact with digital environments without the need for physical input devices. The algorithmic pipeline includes real-time frame acquisition, image pre-processing, hand segmentation, feature extraction, and gesture classification using convolutional and recurrent neural networks. The system supports both static and dynamic gestures, offering flexibility and responsiveness across varied use cases such as media control, presentations, and assistive technology. Robust feedback mechanisms, noise filtering, and performance optimizations ensure accuracy and real-time performance even under variable lighting and background conditions. Comprehensive testing shows high recognition accuracy and low latency, demonstrating the system’s feasibility for deployment on both high-end and resource-constrained platforms. The architecture is modular and scalable, allowing for easy integration with existing software applications and operating systems. Additionally, the system promotes accessibility, enabling hands-free interaction for users with physical impairments. Extensive training on diverse datasets ensures robustness across different hand shapes, orientations, and skin tones. Future work includes expanding the gesture vocabulary, incorporating 3D hand pose estimation, and exploring multi-modal interaction through voice or eye tracking. The proposed gesture interface thus offers a promising alternative to conventional input devices, paving the way for more natural and immersive user experiences.