cageymaru
Fully [H]
- Joined
- Apr 10, 2003
- Messages
- 22,254
20th Century Fox has teamed up with Google Cloud to leverage the power of the Cloud Machine Learning Engine and the TensorFlow deep learning framework to analyze film attributes as data. This allows 20th Century Fox to predict audience preferences in movies. Merlin Video used datasets available from YouTube 8M to break down film into major categories such as male lead. Merlin Video then used labels such as facial_hair, beard, screenshot, chin, human, film and more to analyze films second by second to determine what male audiences prefer in a movie trailer. They found certain cues such as long shots of an object versus intermittent short shorts of the same object could predict customer behavior. This tool was created in a matter of days instead of years and will be used to prototype media plans for marketing campaigns. The research paper can be found here.
With the right infrastructure in place, the team began its analysis on YouTube 8M, a publicly available dataset of YouTube videos. This dataset includes a pre-trained model from Google that is able to analyze specific video features like color, illumination, many types of faces, thousands of objects, and several landscapes. As seen in the figure above, the first step in Merlin's architecture is to parse out these predefined characteristics, as a precursor to determining which elements of the trailer are most predictive of moviegoers' preferences.
With the right infrastructure in place, the team began its analysis on YouTube 8M, a publicly available dataset of YouTube videos. This dataset includes a pre-trained model from Google that is able to analyze specific video features like color, illumination, many types of faces, thousands of objects, and several landscapes. As seen in the figure above, the first step in Merlin's architecture is to parse out these predefined characteristics, as a precursor to determining which elements of the trailer are most predictive of moviegoers' preferences.