Computer Vision-Based Contactless Control Interface for Educational Environments

The relevance of this study is determined by the growing role of distance learning systems and the need to expand interaction modalities with them, particularly in situations of limited access to traditional input devices or within the context of inclusive education. The aim of this study is to design and assess the practical feasibility of a computer vision-based contactless interface for LMS control through operating-system input events, considering robustness under varying lighting, distance, viewpoint, and resolution and comparison with conventional mouse-and-keyboard interaction. To achieve this aim, the following tasks were formulated: to develop a model of a gesture-based LMS control method; to implement both a single-hand gesture mode and a bimanual interaction mode for different control scenarios; to conduct a series of experiments evaluating the accuracy and robustness of gesture recognition under varying recording conditions; and to perform a comparative analysis of the effectiveness of gesture-based control versus the classical “mouse and keyboard” interface. The research methods include computer vision techniques based on hand landmark detection, algorithmic interpretation of gestures using geometric and dynamic features, as well as experimental measurements of recognition accuracy, task execution time, and the number of interaction errors. The evaluation involved five static hand poses, 30 repetitions per pose, and three representative interaction scenarios. The results demonstrate that the proposed system provides stable recognition of large static hand poses across a wide range of lighting conditions and camera distances. Changes in hand orientation were identified as the most critical factor affecting recognition accuracy. All three scenarios were successfully completed in the baseline trials at a resolution of 1280×720, with no critical interaction errors recorded. The practical feasibility of gesture-based control was demonstrated through its integration with a real learning management system, dl.nure.ua, via operating system–level input events. The practical significance lies in its applicability for controlling educational web platforms in distance learning environments, during presentations, and in inclusive education scenarios.