Competitive Analysis System for Theatrical Movie Releases Based on Movie Trailer Deep Video Representation
Authors:
Miguel Campo,
Cheng-Kang Hsieh,
Matt Nickens,
JJ Espinoza,
Abhinav Taliyan,
Julie Rieger,
Jean Ho,
Bettina Sherick
Abstract:
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 h…
▽ More
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.
△ Less
Submitted 12 July, 2018;
originally announced July 2018.
