Original content verification using hash-based video analysis

Abstract

Nowadays, the internet makes it possible for us to upload and share content online. However, the problem is that copying online content has become very easy and has put copyright content at risk. State-of-the-art tools have been designed to detect plagiarised images or texts through the detection of similarities in them. However, there has not yet been a tool for the identification of plagiarised videos which are made up of the fragments of other original videos, that may be legally protected by their authors. This paper presents a tool that has been developed to identify videos created from the fragments of other existing content. The system has been evaluated using videos from the World Cup held in Russia in 2018, some had original content while others were made up of copied fragments. In this way we have been able to verify the feasibility of the system in correctly matching original videos with the plagiarised ones. The results have been satisfactory.