Automatic Video Summarization
Authors: Preeti Vishnu Chandgude, Gayatri Kailas Solanke, Ankita Sudhakar Avhad, Priyanka Ashok Sanap
Full-text Research PDF File: View | Download
Abstract: The primary goal of this system is to propose an improved method for summarising videos in order to present important and concise information to end users. In video, description and keywords are important factors in deciding which video to watch. The proposed method’s main idea is to generate video descriptions automatically. Our approach is critical in reducing the time spent searching for the right video. It aims to save users’ time spent watching unwanted videos by utilising timestamps. One of our approach’s primary goals is actual keyword extraction. Extracted keywords aid in the discovery of videos containing significant video keywords.Frames, emotions, and speech are used to summarise the video. First, the video content is displayed in the frame, followed by a summary of the video content. Second, emotion and how it changes over time are combined with the outputted frame summarization. Third, audio transcription into text takes place, producing an abstractive summary of the audio track. Finally, using natural language processing techniques, the fusion of all summarizations (audio, video, emotion) occurs.
Keywords: Text Summarization, Video Summarization, Latent Semantic Analysis
Paper Id: 230095
Published On: 2023-04-30
Published In: Volume 11, Issue 2, March-April 2023
Cite This: Automatic Video Summarization - Preeti Vishnu Chandgude, Gayatri Kailas Solanke, Ankita Sudhakar Avhad, Priyanka Ashok Sanap - IJIRMPS Volume 11, Issue 2, March-April 2023.