The stage has been set for a new generation of Nielsen ratings. Understanding how much your audience enjoyed the movie goes beyond the volume of tickets sold. Rather now you can track the facial expressions of your audience in real-time to gauge what and which part of the movie was loved by them the most.
Facial recognition system tracks how audiences react to your movies
Caltech and Disney Research are working on real-time tracking of the facial expressions of the audience to measure whether they liked the movie or not, and which was the part they liked the most. In a research paper titled Factorized Variational Autoencoders for Modeling Audience Reactions to Movies presented at IEEE’s Computer Vision and Pattern Recognition conference in Hawaii, demonstrates a new method which can track facial expressions in a theater on a real-time basis. The paper studies nonlinear tensor factorization methods based on deep variational autoencoders to understand that “the latent representation to be learned and the raw data representation is highly complex.”
The researchers collected a large set of face data using an infrared hi-def camera to capture motions and facial expressions of the audience; which was later fed to the neural networks. The research is based on the fact that having a clear understanding of human behavior is essential for building AI systems which exhibit immense behavioral and social intelligence. The basic premise is that people do not explicitly express their problems, and it is on such occasions that body language becomes even more important.
Facial recognition system: Your face says it all!
Facial recognition technology has been a breakthrough in the area of beefing up the security solutions. High adoption of mobile phones, increased instances of identity thefts, the emergence of multimodal biometric systems, and integration of facial recognition in video surveillance, are the chief factors behind the growing demand for facial recognition software systems.
Though industry analysts see the lack of established international standards, as well as high cost of deployment as major roadblocks in the growth of the facial recognition software market – players like Safran, 3M Cogent, and Cognitec Systems, are coming up with innovative solutions to address these issues. Overall, it can be easily concluded that facial recognition technology is here to stay. And, its applications will surely go beyond the standard security aspects.
