Dani Gonzalez

How Computer Vision can Become Mainstream by Providing Real-Time Customer Satisfaction Scores

Facial expression analysis and emotion recognition is a key topic for human-machine and Human-Robot interaction systems, specifically its ability to be processed in real time. Although computer vision has important commercial applications, in the area of emotion processing there’s still a long road ahead, mainly because most available datasets are based on human actors displaying theatrical expressions, with specific lighting, distances and angles. But there is poor generability for wild data, like anything happening in a hotel, a store, or during online activities. Moreover, compared to other Machine Learning hot topics, computer vision has been proven to be very resource greedy and barely optimal or even impractical in real time settings. Robbie has been able to overcome many of these problems: recognizing customer experience from very challenging angles, distances and standard cameras, providing the first automatic customer satisfaction score system at the point of delivery, with wild data and in real time.

Robbie AI is not only able to predict basic emotions, but has developed the first classifiers for non basic emotions, following the circumplex theory of emotion by Dr. James Russell. After getting a totally proprietary dataset of more than 60,000 subjects and 200MM frames and growing, with unique techniques for ground-data combined with monitoring customers in hospitality, Robbie has developed Boredom and Engagement classifiers. Robbie helps retailers, telemedicine platforms, and hospitality managers to analyze instantly problems at the happen, providing emotional business intelligence to improve operations and outcomes. All in real-time, as it happens.

Biography: Daniel Gonzalez is fellow European Union’s Horizon 2020 for science, research and innovation, MIT Launch Mentor, Global Entrepreneur in Residence for UMass Boston, Msc. Computer Scientist at heart but BABSON MBA trained too, and he’s co-founder and CTO of Robbie AI. He’s an expert in real time data streaming engineering, and in his spare time, Dani is defining a totally new way to label wild data for both for training and evaluation. Oh, soccer too.

Buttontwitter Buttonlinkedin
This website uses cookies to ensure you get the best experience. Learn more