Competitive Analysis System for Theatrical Movie Releases Based on Movie Trailer Deep Video Representation
This work addresses audience prediction for movie studios, but it is incremental as it applies existing deep learning and collaborative filtering methods to a new domain-specific dataset.
The authors tackled the problem of audience discovery for movie studios by developing a system that uses deep video representations from trailers and historical attendance data to predict movie attendance and audience profiles. Their hybrid model achieved accurate predictions for existing movies and successfully profiled new movies six to eight months before release.
Audience discovery is an important activity at major movie studios. Deep models that use convolutional networks to extract frame-by-frame features of a movie trailer and represent it in a form that is suitable for prediction are now possible thanks to the availability of pre-built feature extractors trained on large image datasets. Using these pre-built feature extractors, we are able to process hundreds of publicly available movie trailers, extract frame-by-frame low level features (e.g., a face, an object, etc) and create video-level representations. We use the video-level representations to train a hybrid Collaborative Filtering model that combines video features with historical movie attendance records. The trained model not only makes accurate attendance and audience prediction for existing movies, but also successfully profiles new movies six to eight months prior to their release.