Sightcorp logo

Face Analysis Technology

  • Analyze & Identify faces
  • Real-time face analysis
  • Try our software for 2 weeks

Discover Our Products

Use our Face Analysis Technology to analyze faces in videos, images & in real-life environments.

Anonymously detect and measure facial expressions like happiness, sadness, disgust, surprise, anger, and fear with our Face Analysis Technology.

DeepSight Toolkit

  • Easy to use
  • Cross-platform
  • No hard coding needed
  • Connection to Toolkit Visualizer
  • CMS Integration

DeepSight SDK

  • Software Development Kit
  • Track multiple people
  • C++ expertise required
  • Runs offline (one-time internet connection)
  • GPU not required


  • Cloud-based
  • Real-time reports
  • Ready connection with the Toolkit
  • No need to program

Learn more about our Face Analysis solutions

Create engaging and interactive in-store experiences to maximize customer satisfaction

Find The Right Product


We are using and heavily promoting Sightcorp’s Toolkit because of its stability, accuracy and very easy installation and hardware configuration. The real-time aggregated data is immediately accessible through the reporting dashboard, which enables us to scale with ease.

─ S. Ahmedov, CEO OmniChannel


Face Analysis Technology

What is Face Analysis?

Our Face analysis technology analyzes the emotional responses of detected faces. It can recognize 7 basic emotions based on the movement and position of facial muscles. The different analysis produces a confidence score for different types of emotion: anger, disgust, fear, happiness, surprise, sadness, and neutral.

How does Face Analysis Technology work?

Our estimation of the emotional responses will give your insights to optimize and improve the experience of your audience. We can measure the different mood levels during the shopping processes. This technology is often used to monitor people’s reactions after being exposed to content like looking at digital screens.

Which emotions can we analyze?

Our software can identify 7 basic emotions: anger, disgust, fear, happiness, surprise, sadness, and neutral state.

How accurate is the Software?

With an accuracy of 75%,  the different facial expressions can be detected. The accuracy of detecting the overall mood is 92%.

Trusted by Leading Companies

Ready to get started with our face analysis technology?

Test it free for 2 weeks on your favorite desktop platform

Request Free Trial Contact Us

Technical Specifications

The table below shows how FaceMatch SDK performs on the Labelled Faces in the Wild (LFW) dataset:

FPRTPRThreshold (Inverse of distance)
0.10.99900 ±0.002130.55448
0.010.99667 ±0.005370.59791
0.0010.99367 ±0.006050.62989

FPR = False Positive Rate
TPR = True Positive Rate

These results are an indication only and are based on the specific dataset Labelled Faces in the Wild. Customers can expect similar performance, with possible variations due to hardware and the availability of annotated data.