Audience Understanding
Audience Understanding is a multidisciplinary effort—combining computer vision, machine learning, and experimental psychology—to gain deeper insight into how audiences engage with entertainment experiences.
Abstract:
Understanding the audience experience requires not only an appreciation of how emotional states manifest in behavior—such as body motion, facial expressions, and patterns of eye movement—but also a comprehensive view of how content is produced with an intent to induce certain emotional states. Narrative experiences aim to take audiences on an emotional journey. Our research is focused on understanding whether that intended journey is being taken. Our approach uses cutting-edge computer-vision methods for interpreting the behavior of volunteer audiences and machine-learning models for analyzing content for its structure and creative intent. The result is a tool for both executives and creatives that allows for a better moment-to-moment understanding of how audiences relate to the Disney experience.
Tech Transfer:
Insights from the Audience Understanding project are being used by the Disney-ABC Television Group to inform content decisions.
Publication Highlights
Harnessing Object and Scene Semantics for Large-Scale Video Understanding
June 27, 2016
IEEE Conference on Computer Vision Pattern Recognition (CVPR) 2016
Zuxuan Wu (Fudan University) Yanwei Fu (Disney Research) Yu-Gang Jiang (Fudan University) Leonid Sigal (Disney Research)
Learning Activity Progression in LSTMs for Activity Detection and Early Detection
June 27, 2016
IEEE Conference on Computer Vision Pattern Recognition (CVPR) 2016
Shugao Ma (Boston University) Leonid Sigal (Disney Research) Stan Sclaroff (Boston University)
Predicting Movie Ratings from Audience Behaviors
March 24, 2016
IEEE Winter Conference on Applications of Computer Vision
Rajitha Navarathna (Disney Research/Queensland University of Technology) Patrick Lucey (Disney Research) Peter Carr (Disney Research) Elizabeth Carter (Carnegie Mellon University) Sridha Sridharan (Queensland University of Technology) Iain Matthews (Disney Research)
Space-Time Tree Ensemble for Action Recognition
June 7, 2015
IEEE Conference on Computer Vision Pattern Recognition (CVPR) 2015
Shugao Ma (Boston University) Leonid Sigal (Disney Research) Stan Sclaroff (Boston University)
Jointly Summarizing Large-Scale Web Images and Videos for the Storyline Reconstruction
June 23, 2014
IEEE Conference on Computer Vision Pattern Recognition (CVPR) 2014
Gunhee Kim (Disney Research) Leonid Sigal (Disney Research) Eric P. Xing (Carnegie Mellon University)
Poselet Key-framing: A Model for Human Activity Recognition
June 23, 2013
IEEE Conference on Computer Vision Pattern Recognition (CVPR) 2013
Michalis Raptis (Disney Research) Leonid Sigal (Disney Research)