How many landmarks do you track?
Our face tracking technology is not built on landmarks. Instead of landmarks, we recognize 37 face characteristics represented as morphs for the default face mesh. This is the most optimal number that allows us to get all the needed data to detect and track a face in the camera.
When demoing our technology, you may see 87 landmarks marked by which you can evaluate the accuracy of our face tracking.
How does your technology differ from competitors?
Since we don’t track landmarks but recognize only 37 face parameters, we achieve a much faster technology performance and better accuracy. The algorithms need to process only 37 numbers saving the capacity resources.
Patented anti-jitter. As we significantly save on the capacity resources we are able to run algorithms several times within one frame. It shows us the constant data (face itself) and the variable one (the noise). We detect and process the noises within one frame due to which the technology provides no lag.
Our face tracking technology is optimized specifically for mobile to function equally well on iPhone-Xs and $100-Androids, at low lighting, and with up to 70% occlusion.
Our device support starts with iPhone 5S, Android 5.0 with Camera 2 API and Open GL ES 3.0 which covers 97% of all iOS devices and 80% of all Android ones.
Are your models trained for all people skin tones?
Yes, we use the balanced datasets that include people's faces of different ages, genders and nationalities.