Skip to main content

Applying AI Techniques to Auto-categorise Violent Images

Supervisors

Harjinder Lallie
(http://www2.warwick.ac.uk/fac/sci/wmg/people/profile/?wmgid=856 http://www2.warwick.ac.uk/fac/sci/wmg/people/profile/?wmgid=856)

Suitable for

MSc in Computer Science
Mathematics and Computer Science, Part C
Computer Science and Philosophy, Part C
Computer Science, Part B
Computer Science, Part C

Abstract

"The analysis of violent images in digital forensics cases can take a long time and be traumatising for digital investigators. Artificial intelligence (AI) techniques can aid in automating the categorisation of violent images, and datasets such as the Hollywood violent scenes dataset (https://www.technicolor.com/dream/research-innovation/violent-scenes-dataset) can be used to aid the categorisation. This project involves developing a tool which uses AI techniques to categorise violent images. Please note that you will not be required to look at 'real' violent images, the images in the dataset referenced herein are from Hollywood films such as Reservoir Dogs, Armageddon, Harry Potter and the Order of the Phoenix, Billy Elliot, Kill Bill 1 and The Wizard of Oz Useful reference(s): C.H. Demarty,C. Penet, M. Soleymani, G. Gravier. VSD, a public dataset for the detection of violent scenes in movies: design, annotation, analysis and evaluation. In Multimedia Tools and Applications, May 2014. "

Prerequisites: suitable for someone who has some AI experience. Knowledge/experience of digital forensics is not critical