Movies data specifically has become one of the most entertaining sources for viewers. However, browsing a movie in enormous collections and searching for a desired scene within a complete movie is a tedious and time-consuming task. A PhD position is available at the Centre for Visual Computing on the topic of movie indexing using computational methods and techniques. In particular, this research project will focus on the development of efficient methods for movie summarization (trailer generation), which will play a pre-processing step for efficient indexing of movies in large archives.
This project aims to develop Visual Computing and Artificial Intelligence-Based movies data management system by producing a summarized video from the movie, which contains the most important events, and scenes. Hence, the viewers may have an idea about the context and the semantics of the movie by watching only the semantically relevant generated trailers.
Our research group has an established track record in developing video summarization, content-based information retrieval, indexing technologies and various other visual computing frameworks. This project will deliver, a new computer vision system for movie indexing and summarization. This research project will have much impact upon the development of new video indexing tools for the entertainment industry. Research in this areas is very much in collaboration with Engineering and Media.
Students having background in image processing with expertise in any of the programming languages (Matlab, Python, C, C++, C#) are welcome. The domain of this project can be adjusted as per the qualification and interests of students.