Unsupervised Video Summarization Using Deep Learning Approach with Deep Feature Semantics

One of the potential methods for efficiently understanding video material is video summarization, which does so by picking out relevant scenes from the movie scene. As the content of videos varies widely, making video summarizing considerably more difficult as prior knowledge is nearly inaccessible, the goal here is to provide a video summarization that is engaging to the user and accurately represents the context of content using an unsupervised learning approach. This paper offers a deep learning solution to tackle this issue in which deep video features represent many layers of content semantics, such as actions, scenes, and objects, to boost the performance of baseline video summarizing methods. The deep features are extracted scenes taken directly from the original movie, and an unsupervised k-means clustering-based algorithm is applied to summarize the video. Our video summaries are evaluated against reference approaches and state-of-the-art models on two standard datasets.