IBM Research Deploys Machine Learning for 'Morgan' Trailer

Using machine learning techniques and experimental Watson APIs, the IBM research team trained a system using trailers for 100 horror movies and segmenting out each scene.
Author:
Publish date:
Updated on

Twentieth Century Fox partnered with IBM Research to explore the use of artificial intelligence to create a trailer for the suspense/horror film Morgan that would keep audiences on the edge of their seats.

“Our team was faced with the challenge of not only teaching a system to understand ‘What is scary?’ but then to create a trailer that would be considered ‘frightening and suspenseful’ by a majority of viewers,” explains IBM Research’s John R. Smith.

Using machine learning techniques and experimental Watson APIs, the IBM research team trained a system using trailers for 100 horror movies and segmenting out each scene. Once each trailer was segmented into “moments,” the system completed a visual analysis, an audio analysis and an analysis of each scene’s composition so it would “understand” the types of scenes that fit into the structure of a suspense/horror movie trailer.

“We [then] fed the system the full-length feature film Morgan,” Smith says. “After the system ‘watched’ the movie, it identified 10 moments that would be the best candidates for a trailer. Our system could select the moments, but it’s not an editor. We asked a resident IBM filmmaker to arrange and edit each of the moments together into a comprehensive trailer.

“The combination of machine intelligence and human expertise is a powerful one,” Smith adds. “This research investigation is simply the first of many into what we hope will be a promising area of machine and human creativity.”

Watch the completed trailer below.

Related