AI Techniques in Video editing
Video production and distribution have become very affordable and accessible. A large body of research is available in AI techniques for audio, visual and language processing and, more recently, in the generation of multimedia content. These AI techniques offer opportunities to design better video workflows by means of automating trivial tasks, offering creative suggestions, other modes than frame by frame video manipulation, augmenting the videos clips with data for search across the video library as well as within each videos - all towards the goal of offering better video workflows.
Though AI techniques are promising as design materials for video production, the usage of AI in video production tools is very limited. There is lack of studies and expertise in how video editors would perceive and use machine learning in their work. As a part of this ongoing joint innovation project between university and industry, this (sub? PhD ?) project aims to explore the the challenges of integrating AI techniques into video editing workflows. Through developing various functioning AI embedded prototypes for video production and using them to run studies, we aim to explore the design space of using AI in video editing interfaces and the potential of human-AI collaboration in creative designs. In addition to academic contributions in HCI, the project also aims to contribute some of the results back into the product development of the company.