Development of a System for Automatically Output a Title Using Face Detection

In this final qualifying work, a system is being designed and created to automate the process of displaying a title with information about a person in the frame during online broadcasts, created on the basis of a client-server architecture and neural networks and capable of recognizing people's faces and generating an NDI video stream with subtitles in real time. The developed application will automate the work of directors and cameramen when conducting online broadcasts with a large number of participants: conferences, lectures and meetings. The result of the work is a software solution that will reduce the time spent on preparing for the filming process and simplify the routine tasks of the film crew, especially for small studios with limited staff, as well as reduce the number of errors in the event of unforeseen situations, for example, changing the order of speakers or changing their number. The volume of the final qualifying work is 55 pages. The work contains 15 illustrations, 2 tables, 1 appendix. The bibliographic list consists of 48 sources.