Trump Backs Nexstar–Tegna Merger Amid Shifting U.S. Media Landscape 
American Airlines CEO to Meet Pilots Union Amid Storm Response and Financial Concerns 
SoftBank Shares Slide After Arm Earnings Miss Fuels Tech Stock Sell-Off 
OpenAI Expands Enterprise AI Strategy With Major Hiring Push Ahead of New Business Offering 
Amazon Stock Rebounds After Earnings as $200B Capex Plan Sparks AI Spending Debate 
Global PC Makers Eye Chinese Memory Chip Suppliers Amid Ongoing Supply Crunch 
Sony Q3 Profit Jumps on Gaming and Image Sensors, Full-Year Outlook Raised 
TSMC Eyes 3nm Chip Production in Japan with $17 Billion Kumamoto Investment 
Rio Tinto Shares Hit Record High After Ending Glencore Merger Talks 
TrumpRx Website Launches to Offer Discounted Prescription Drugs for Cash-Paying Americans 
Ford and Geely Explore Strategic Manufacturing Partnership in Europe 
SpaceX Pushes for Early Stock Index Inclusion Ahead of Potential Record-Breaking IPO 
CK Hutchison Launches Arbitration After Panama Court Revokes Canal Port Licences 
Tencent Shares Slide After WeChat Restricts YuanBao AI Promotional Links 
FDA Targets Hims & Hers Over $49 Weight-Loss Pill, Raising Legal and Safety Concerns 
Nvidia, ByteDance, and the U.S.-China AI Chip Standoff Over H200 Exports 
Instagram Outage Disrupts Thousands of U.S. Users 
PALO ALTO, Calif., March 02, 2017 -- Sighthound, Inc. today announced the release of a deeply learned emotion recognizer on its Sighthound Cloud API service. The emotion recognizer analyzes an image or video stream to find all faces and determine, for each face, the emotion along with an associated confidence level. Sighthound’s emotion recognizer has been trained to recognize seven different types of emotions.
Sighthound now has the world’s most accurate age detector, gender recognizer and emotion classifier, providing significantly more accurate results than the Microsoft Cognitive Services API.
Real Age Estimation
| Methods | Mean Absolute Error |
| Sighthound | 5.76 |
| Rothe et al. | 7.34 |
| Microsoft | 7.62 |
| Face++ | 11.04 |
Emotion Recognition Accuracy
| Methods | Accuracy | |
| Sighthound | 76.1% | |
| Microsoft | 61.3% | |
Gender Recognition Accuracy
| Methods | Accuracy | |
| Sighthound | 91.00 | % |
| Microsoft | 90.86 | % |
| Rothe et al. | 88.75 | % |
| Levi and Hassner | 86.80 | % |
| Kairos | 84.66 | % |
| Face++ | 83.04 | % |
Full benchmark results with academic references and white paper are available here:
https://www.sighthound.com/technology/age-gender-emotion/benchmarks
Alongside the developer API, Sighthound released a demonstration version of the emotion recognizer tool on its website today for testing.
Sighthound makes it easy for developers to get started by offering free tutorials, sample code, and documentation to help integrate its APIs into their own applications.
Developers can use the free Sighthound Cloud API at https://www.sighthound.com/products/cloud. Paid accounts (Basic, PRO, and Enterprise) have access to production servers for commercial use and begin at $49 per month.
About Sighthound
Sighthound is the easiest way to add state-of-the-art computer vision software to your products and services. www.sighthound.com
Media Contacts: Sighthound: Rainer Flor, Sighthound, [email protected]
