Amazon AI ‘Rekognition’ Can Now Detect Fear

Amazon is way more than just online shopping. They have their hand in things like analytics, blockchain, and robotics. Amazon Web Services (AWS), which provides on-demand cloud computing platforms, generated over 6.1 billion dollars in revenue last year. They also offer another service to some of their customers…facial recognition.

Amazon’s Rekognition is an application which analyzes images and video, identifying objects, people, text, scenes, and activities, as well as noticing any inappropriate content. It can also detect seven emotions including ‘Happy’, ‘Sad’, ‘Angry’, ‘Surprised’, ‘Disgusted’, ‘Calm’ and ‘Confused’.  They’ve recently added an eight emotion: ‘Fear’. So they’ll apparently know when a product is freaking you out.

Amazon Rekognition Facial Recognition - YellRobot
credit: Amazon

Amazon Rekognition Can Determine Age, Gender, and Emotion

A facial analysis algorithm generates metadata from detected faces in an image. It will attempt to determine things like emotion, gender, age, face pose, face image quality, and face landmarks. According to the company, Rekognition can detect, analyze, and compare faces for a wide variety of uses such as user verification, people counting, and public safety.

The algorithm is continually trained on new data to expand its ability to recognize objects, scenes, and activities. Along with adding the ability to detect ‘Fear’, Amazon has improved age, gender, and emotion detection accuracy.

“With this release, we have further improved the accuracy of gender identification. In addition, we have improved accuracy for emotion detection (for all 7 emotions: ‘Happy’, ‘Sad’, ‘Angry’, ‘Surprised’, ‘Disgusted’, ‘Calm’ and ‘Confused’) and added a new emotion: ‘Fear’. Lastly, we have improved age range estimation accuracy; you also get narrower age ranges across most age groups,” the company said in a blog posting.

Check out our articles on AI helping to detect dangerous driving and Gait recognition detecting mood by the way you walk.