Using AI & Movie Trailers to Predict Audience
Subscribe:  iCal  |  Google Calendar
Las Vegas US   8, Jan — 12, Jan
Zürich CH   31, Jan — 4, Feb
Leamington Spa GB   31, Jan — 3, Feb
Bradford GB   6, Feb — 11, Feb
Bradford GB   7, Feb — 9, Feb
Latest comments
by Vitaly Lashkov
3 min ago

Ok, what about LODs and billboards in "most commonly used method"..? In comparing, Most commonly used method can help me to reach less synthetic look. And with progressive LODs keep overdraw on predictable level. Brutforce approach never works - confirmed)

честно сказать я в ахуе

finally some good news

Using AI & Movie Trailers to Predict Audience
26 July, 2018

NVIDIA has shared a nice article that shows the way 20th Century Fox uses deep learning to predict a film’s audience from trailers. The system is said to be trained on historical movie attendance records to “find non-trivial associations between the video trailer features, and future audiences choices after the movie releases in theatres or on streaming services”. The whole analysis focuses on extracting features like “color, illumination, faces, objects, and landscapes” to predict accurate attendance for existing movies, and the team plans to use for future movies.

“Video trailers are the single most critical element of the marketing campaigns for new films,” the Fox researchers stated in their paper. “They increase awareness among the general moviegoer population, communicate the plot of the movie, present the main characters, and reveal important hints about the story, the tone, and the cinematographic choices.”

Using NVIDIA Tesla P100 GPUs on the Google Cloud, with the cuDNN-accelerated TensorFlow deep learning framework, the researchers trained their convolutional neural network on hundreds of movie trailers released over the last years, as well as millions of attendance records.

For inference, the team uses the same GPUs used during training.

“By finding a suitable representation of these features, and by feeding them to a model that has access to historical movie attendance records, it is possible to find non-trivial associations between the video trailer features, and future audiences choices after the movie releases in theaters or on streaming services,” the researchers stated.

The neural network has the potential to assist movie producers and executives make real-world decisions at different stages of a marketing campaign, the researchers said.


You can learn more here or read the full paper here


Leave a Reply

2 Comments on "Using AI & Movie Trailers to Predict Audience"


I love watching movies a lot. Now its easy to watch all movies, TV shows and series for free on Android using Cinema apk. Get it from


This is a great technology for film makers like. Thanks.