Comparative Analysis of Text-based Video Summarization Techniques using Deep Learning

Rapid advancements in video technology, as well as the widespread availability of video capturing equipment, generate large amounts of data that must be managed efficiently in order to improve usability. Automatic video summarizing is a method of navigating through large amounts of data and producing a short summary that can be used in a variety of applications. Traditional video summary, on the other hand, creates concise videos with no discernible qualitative or quantitative loss of content. It just considers the most significant aspects of the video while providing a summary that is independent of the user’s interest. Because video summarizing is a highly subjective activity that cannot be performed by traditional approaches, people are more interested in customized summaries. The solution to this challenge is multi-model video summarization, which aids in the production of user-interested summaries by taking into account a variety of factors. This paper discusses the different approaches used in query based i.e. text-based video summarization and enlighten the pros and cons of existing methods to help research community to work in this direction.