Facial recognition allows a computer system to verify a person’s identity. Recognition already requires the detection of the position of a face in a photo or video stream. Then this face is transformed into a vector of 128 or 2048 values depending on the used model, representing each of the characteristics. The decision on a person’s identity is made according to the distance of the recognized face from the faces pre-registered in the database.
We have the necessary software building blocks to add detection and/or facial recognition to your system, including embedded systems with limited computational constraints.
The human being uses gestures in an infinite number of contexts to communicate but also to control a functioning. The latest image processing technologies now make it possible to recognize and monitor the position of the hand in a space and to recognize the posture of the hand in real time. Unlike existing industrial solutions, we allow you to recognize gestures in a more generalist way adapted and completely optimized for your particular use.
We have been working on the recognition of human gestures since 2009 through 2D touch interfaces, then 3D from 2011. We then began research on the different aspects related to the generation of gestures to control interfaces.
The subsequent detection and tracking of objects and shapes is usually the first part of a large system. We help you to set up this software brick by connecting it to your system. Some of our clients have used our work to detect certain objects and then track their movements by turning the cameras/projectors. Automatic segmentation operations can be added after which other aspects can be detected.