The Image class provides an interface to work with images and textures. The individual pixels of the image can be read or changed by the simulation as desired.

The image class can read rendered images from the screen using the method read-pixels, so that agents in the 3D world can have access to real rendered data. In addition, the method get-pixel-pointer can be used to provide a pointer to the RGBA pixel data so that plugins can access and analyze image data. This could be used, among other things, to implement agent vision.