CVJan 21, 2021

Video Summarization: Study of various techniques

arXiv:2101.08434v11 citations
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of efficiently summarizing videos for users in domains such as media and surveillance, but it is incremental as it reviews existing methods without introducing new approaches.

The paper presents a comparative study of various techniques for video summarization, focusing on converting lengthy videos into shorter ones that capture important events, with results and architectures discussed for different contexts like sports and documentaries.

A comparative study of various techniques which can be used for summarization of Videos i.e. Video to Video conversion is presented along with respective architecture, results, strengths and shortcomings. In all approaches, a lengthy video is converted into a shorter video which aims to capture all important events that are present in the original video. The definition of 'important event' may vary according to the context, such as a sports video and a documentary may have different events which are classified as important.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes