The possibilities of modern cognitive computing services are endless. That's why we are really excited about services like text, speech and image recognition. 

Because we believe that AI will have an even bigger impact in our lives than the smartphone revolution has had, we explore these services on a daily basis.

Overwhelmed by the possibilities, we developed our own face recognition demo. When entering our head office in Amsterdam, visitors will notice a big screen with a camera. Once a visitor’s face has been detected, further analysis takes place using cloud face recognition services based on the power of machine- and deep learning. The service provides insights like age, gender, emotion and other facial features like reading glasses, beard, moustache, lip & eye makeup. The demo also has a look-a-like detection feature. This appliance shows a person you look most similar to if there is more than a 50% match. A very fun and appealing demo.

 

A nice feature of this prototype is that it gives users a near real-time experience and it is even capable of recognizing our visitors by name. With a set of images (profile pictures) current cloud services - like Microsoft Azure Face API - can detect faces very accurate. Mistakes only seem to happen with bad quality images.

The demo shows the true power of today’s cognitive cloud services powered by AI and can easily be enriched with your own data. Several use-cases can be thought of such as supporting your employees with relevant data, like their next appointment. Attracting visitors to your booth at an event is another great example. And because the system now is capable of logging statistical data, you could provide insights like the happiness level per day, top visitors, average age, et cetera. We made a real-time dashboard using Qlik Sense.

Together with NVM we launched the magic mirror for real estate agents (De MakelaarSpiegel). This application recognizes real estate agents and shows insights like age, gender, emotion and other facial features. Because the application is linked to the NVM database, the magic mirror also shows the last three objects the real estate agent has sold. The magic mirror is a fun and personal way to introduce NVM real estate agents to the latest technologies.

The amount of cloud cognitive services on offer, grows rapidly – Microsoft for example, is already working on broadening and deepening its portfolio. Interested into a summary of relevant cognitive services, how we deploy these services and how the performance of these services of various different providers compare? Download our research. 

The necessary technology for face recognition is very easy accessible nowadays. However, you can't just implement it like that. Privacy legislation states that you can't process personal data without permission. And making or processing a facial photo for face recognition is processing of personal data according to the Dutch Data Protection Authority (Autoriteit Persoonsgegevens). The DPA also indicates that face recognition by means of cameras could fall under camera surveillance, and this makes it prohibited in public areas. The EU's new privacy legislation (GDPR) sets strict requirements for the processing of biometric data (like facial photos). However, to what extent this legislation applies to face recognition is not yet clear. The EU's new ePrivacy Regulation will provide more clarity, but is still in the making. Until then it seems wise to inform users in advance when applying face recognition in public areas. A lot is possible with permission, but your specific situation is decisive. So, it is important to get legal advise for your specific case.

 

Our face and image recognition demos demonstrate that cloud AI services are becoming very mature and - with this technological foundation - you could build cases for all industries.

 

Interested? Drop us a line and get in touch with Peter Bakker.