Using our eyes, we are able to perceive our environment in the form of images. This is facilitated by cells located on the retina of the eyes that react sensitively to light. The cones of the retina are specialised cells that react to light intensity and colours, while the so-called rods react to white and black and the various gradations of grey tones. The sensory cells transmit the data to the brain which, based on this information, deduces an object’s form, colour, light intensity and speed of movement.

To ensure a sharp image of the viewed object, the muscles in the iris narrow or widen the pupil to regulate the amount of light entering the eye; and the lens alters its curvature according to the distance to the object being viewed. As a rule, this occurs automatically and unconsciously. The rays of light that come through the lens produce an image on the retina that is dispersed into many individual dots. Each dot corresponds to a cone or a rod. The impulses generated in these cells are transmitted via the optic nerve to the visual centre of the two halves of the brain, which then supplies the overall image of the object being viewed. And since we have two eyes, we are capable of seeing in three dimensions and judging an object’s distance.

Object recognition

The ability to recognise objects seems so trivial to us that we are hardly even aware of it. We recognise friends even with new glasses, make out individual persons within a group and estimate the relative distances between them. From afar, we recognise a tree and, upon closer inspection, we are able to classify it as an apple tree.


We are also able to recognise patterns, even when the concrete expressions of the pattern strongly vary.

«Gazers» serve as Ada’s eyes

The movable directional cameras (gazers) enable Ada to classify motion signals and colour histograms of hands and faces. What Ada finds interesting, for instance, are persons exhibiting a high light value (such as light-coloured clothing), fast-moving individuals and those who are in motion relative to other visitors. Ada is also able to resolve a group of persons into individual persons. If Ada finds a person interesting, then she can follow the individual with her movable directional cameras and take close-ups. While Ada does not recognise faces, she is able to differentiate among individuals, such as by the colour of their clothing. Ada will also be able to recognise and accordingly interpret simple gestures like waving one’s hand.