Краткое описание

The paper is devoted to solving the problem of preparing educational video content without an operator. The paper describes the development of algorithms and scenarios for identification, localization, recognition and classification elements in a frame, including using neural network technologies. The developed complex of algorithms and scenarios is implemented in the software system of intellectual analysis of educational video lectures, real-time temporal and sense synchronization of video lectures and presentation materials. It is assumed that the use of the developed system will lead to a significant acceleration and reduction in the cost of the preparing educational video lectures process.