One of the most basic tasks associated with facial recognition is detection of frontal or near-frontal faces in a photo, regardless of orientation, lighting conditions or skin color. For each face, you get a complete analysis of key points (landmarks) around eyes, eye brows, jaw, nose and mouth. This feature answers the question "Where are the faces in this picture?"


Recognition or identification involves confirming someone’s identity, once their face has been detected within the image, by searching through hundreds of thousands of known faces in less than one second. Each match is given a score, so that you can easily sift through the response and make an informed decision. This feature answers the question "Who are the people in this picture?"


Detected faces are classified depending on age, gender, ethnicity or emotion, so this feature is especially useful for retail analytics and designing digital signage with real-time targeting. Classification doesn’t seek to single out individual identities, but rather to employ built-in analytics in order to generate the demographic make-up of people in a certain area.


Want to find out which actor or model you look like most? Are you curious if there’s someone else in the company who kinda looks like you? Our API can answer all those questions by searching for doppelgangers and lookalikes in a pre-existing set of faces. This feature answers the question "Who does Mary look like?"


Our API can compare two or more faces in order to assess if they belong to the same person - even if that person is unknown. No more manually comparing selfies with photo from identity card. This feature answers the question "Is it the same person in all these pictures?"


Counting is closely linked with classification, in that it allows you to know exactly how many people have been detected within a given time frame. It works on both camera streams and groups of photos and doesn’t require identifying a specific person. The measure is simply statistical.


Returning or re-identification tells you how many times a person reappears after being identified once. It can be especially useful for smart surveillance, but also applies to retail and hospitality, especially when one of your goals is identifying and rewarding loyal customers.

Liveliness Detection (beta)

For authentication and fraud prevention scenarios, having a match between two faces is sometimes not enough, especially when pictures are fed in from the user’s device. As a premium feature for use cases where security is key, VisageCloud can analyze a burst of frames (several pictures taken in quick succession) to assess whether they legitimately depict the subject or if they depict a photograph of the subject or a device screen. This feature answers the question "Is this a legitimate picture or a fake?"

Contact us

Let us explore together how VisageCloud can best work for your use case


P: +40.724.714.234

Bucharest, Romania

© 2018 VisageCloud