Improved Classification System Surpasses Major Technology Providers and Academic Benchmarks For Facial Coding
NEW YORK, April 28, 2020 (GLOBE NEWSWIRE) -- Realeyes, a leading computer vision and emotion AI company, announced today the availability of its next-generation facial coding technology. Realeyes uses front-facing cameras and the latest in computer vision and machine learning technologies to detect attention and emotion among opt-in audiences as they watch video content. The enhanced classification system will provide customers with more sensitive, accurate insights into the emotional impact of their video content.
Realeyes continues to set the industry standard for facial coding accuracy. The improved classification system results in a 20% increase in emotion detection across all measured emotions from facial cues. It also reduces occasional false positive emotion readings by half. Realeyes is the most accurate emotion detection technology among leading API cloud providers, based on an internal benchmark study of thousands of videos.
“Our technology has reached a new level of sophistication, with the accuracy of our detection beginning to rival that of humans across certain emotions like happiness and surprise,” said Elnar Hajiyev, Chief Technology Officer at Realeyes. “Realeyes is building transformational apps to enable companies to create more remarkable experiences for people, and it starts with a foundation of world-class core vision technology. Realeyes today holds 11 patents covering different aspects of building emotion AI technology, and has 29 pending.”
The updated classifications are applied to all Realeyes products and improve on the platform’s ability to accurately analyze a wider variety of viewers’ faces and emotions. The new classifications allow for more nuanced reading through more sensitive emotional curves and bring greater value to the data collected through facial coding.
Said Hajiyev: “More accurate detection enables companies to better understand the pure attention and emotion response of their audiences. However, more accurate emotion detection also enables advertisers to better predict in-market outcomes such as video view-through rates, so they can create more engaging creative to maximize media spend.”
Realeyes’ upgraded classifiers allow for a greater range of facial measurements across ethnicities, especially those of Asian heritage. Combined with improvements to Realeyes’ performance on mobile devices, the updates pave the way for an entirely new range of products and applications based on emotion AI, along with relevance in new markets around the world. Realeyes announced the appointment of its Japan country manager Kyoko Tanaka followed by last year’s strategic investment from notable international investors Draper Esprit, and NTT DOCOMO Ventures, Inc., the VC arm of NTT Group, Japan’s leading mobile operator.
Trained on the world’s richest database of facial coding data, Realeyes’ technology now incorporates more than 615 million emotional labels across more than 3.8 million video sessions to provide more nuanced insights into the emotional impact of video content. The recent update strengthens Realeyes’ predictive modeling for behaviors like view-through rate and responses like interest and likability, while providing best-in-class results 8x faster than its previous version.
Using front-facing cameras and the latest in computer vision and machine learning technologies, Realeyes measures how people feel as they watch video content online, enabling brands, agencies and media companies to inform and optimize their content as well as target their videos at the right audiences. Realeyes’ technology applies facial coding to predictive, big-data analytics, driving bottom-line business outcomes for brands and publishers.
Founded in 2007, Realeyes has offices in New York, London, Tokyo and Budapest. Customers include brands such as Mars Inc, AT&T, Hershey’s and Coca-Cola, agencies Ipsos, MarketCast and Publicis, and media companies such as Warner Media and Teads.
(917) 826 - 1103