A Systematic Review of the Literature in Dynamic Video Summarization

Abstract

Due to the increase of video data generated in today's life, finding ways to mine these data or summarize them has become a great challenge. Thus, the main objective of this work is to investigate why video summarization is needed and how it could be done in the best possible way. To meet this goal, a systematic literature review (SLR) was conducted on articles published from 2020 to 2021 to identify the primary studies on the use of video skimming in data mining, following a predefined review protocol. The contributions and the types of evidence reported regarding the benefits of this method were also examined. In summary, the main findings of this work are: (1) the use of video summarization is widely variant, and more new domains are being introduced each year, (3) industrial researchers have not participated in academic studies, and (4) different evaluation metrics and datasets have been introduced, and studies are attempting to introduce new methods. Finally, this work showed several promising research opportunities that are pretty important and interesting but underexplored in current research and practice.