Cydral FastFace is a comprehensive facial recognition system capable of finding identical or similar persons. Based on advanced AI algorithms, Cydral Technology's experts in machine learning createda smart and innovative search engine to enhance the user experience for people identifying.
Given a detected face extracted from a picture taken by a camera or loaded from a library and a set of public candidate faces to search with, the FastFace search engine finds a small set of facesthat look most similar to the target face. Additionally, and for each uploaded image, FastFace returns additional attributs such as the estimated age and the gender.
Unlike other face recognition technology, our FastFace system allows quick searches in big data sets. Our technology can work with any photographic database, current results though it use publicimages on the Internet indexed by our specific crawlers.
Typical Use Cases: ✓ Large-scale search by photo: you see a celebrity on the street that you can't quite recognize... FastFace him/her! ✓ "Who is this actor?": try out the name suggestion for well-known celebrities during a movie ✓ Person identification: identify a person on photo in few seconds with an impressive precision ✓ Gender and age-guessing tool: besides finding your look-alike, our system will do its best to guess the age and gender of the people in the shot, a feature also to maximize your efficiency ✓ Easy to use and funny: simply select the image you want to use in your phone gallery or take a photo to instantly start your search ✓ Advanced search engine: our flexible search engine behind the mobile application provides effective feature filtering to restrict results
Tips to improve your search experience: ✓ Some things can trip up the search process, such as bad image quality, sometimes objects that obscure the face or a face that’s mostly in profile ✓ To improve the search accuracy, it is sometimes valuable to directly upload photos rather than using the mobile phone camera ✓ On some devices, an image capture using the camera intent gets rotated. In this case, the accuracy drastically decreases