Sightcorp logo

Face Reader Online

Everything about Face Reader Online

     

What is an Online Face Reader?

An Online Face Reader enables you to conduct facial analysis online, or to integrate face reading
capability into another online application. The image processing then usually takes place in the cloud,
with results being returned immediately.
Sightcorp’s Online Face Reader technology, CrowdSight F.A.C.E. API, for example, is a cloud-based web service that allows you to integrate facial analysis into any online application. With F.A.C.E. API, you can incorporate online face reading into your own apps without having to program and without having to acquire powerful hardware.

When to use an Online Face Reader?

An Online Face Reader is useful when you want to analyze faces and facial expressions remotely.
Here are some examples:

  • When you want to measure the emotional response to online ads and other online content (such as games and movie trailers, for example), and you want to do so remotely.
  • When you want to personalize app or website content and ads based on the demographics of the audience in real time.
  • When you want to conduct user experience research and gain consumer behavior insights remotely.
  • When you want to integrate face reading capability into another online application.

 

When to use an Offline Face Reader instead?

  • When you don’t need to capture image or video footage from a remote location. For example, when you are capturing footage in a retail store or in the context of digital signage, you can easily use face reading software on a local device.
  • When you want to integrate face reading capability into an offline application.
  • When you want to be in full control of the data you are capturing and processing.

 
At Sightcorp, we offer the following face reading solutions with offline capability:

 

What can an Online Face Reader do?

Generally speaking, an Online Face Reader will have the same or similar functionality as face reading
software that can be used offline. This means that it will be able to:

  • Detect faces
  • Recognize emotions such as happiness, sadness, anger, fear, disgust, surprise, and contempt
  • Determine demographic information, such as age and gender

 
One of the main differences is that with an Online Face Reader, data will not be processed locally.
This means that you may need to pay extra attention to any data security and privacy concerns.
Click here to discover Sightcorp’s Online Face Reading technology

Here are other articles you might find interesting:

   

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.