Support Support Home » Vision Libraries » ThirdSight



Sightcorp is an AI spin-off from the University of Amsterdam (UvA), specializing in easy-to-use face analysis and face recognition software.
We transform computer vision and deep learning research into commercial solutions that give our customers accurate, real and actionable insights into their audiences.
Our audience analytics software can anonymously analyze faces in images, videos, and real-life environments, while our face recognition software assists globally in accurately verifying identities and authenticating users in fields such as Access Control, KYC, and Payment Verification.

Sightcorp offers two software development toolkits, CrowdSight and InSight:

CrowdSight SDK

Designed for Crowd Analysis
CrowdSight SDK is a crowd analysis software which allows to analyse audiences or passers-by behaving spontaneously in their natural environments. Retrieve relevant information such as facial expressions, head pose, clothing style, age and gender in real time and create interactive and targeted content based on these insights.

CrowdSight SDK can collect data in real time on:
  • Age
  • Gender
  • Mood (positive, negative)
  • 6 facial expressions (happiness, sadness, fear, surprise, anger, disgust)
  • Foot traffic in front of signage
  • Attention time
  • Head pose
  • Eye position
  • Clothing style
  • Face verification

Insight SDK

Designed for Accurate Individual Analysis, 70cm from the camera
InSight SDK accurately tracks facial muscles and measures facial expressions, eye tracking and head gaze tracking of individuals. Understand how users react to different stimuli and which areas they pay attention to on a screen. Improve the level of interaction with your devices and make your content more effective based on these insights.
InSight SDK can only be used in situations where the user is close to the camera and can perform a calibration procedure (collaborative users).
After calibration, the SDK can collect data with state-of-the-art accuracy on:
  • Age
  • Gender
  • 3D facial deformations.
  • 7 facial expressions (neutral, happiness, sadness, fear, surprise, anger, disgust)
  • Eye gaze tracking
  • Head gaze tracking


Both SDKs allow software developers to integrate this powerful technology in your facial analysis and digital signage solutions using the C++ programming language. The technology is based on the OpenCV image processing library, which is responsible for the integration of our cameras using the HighGUI CvCapture class.

More info

Additional information can be found at the Sightcorp website.