Carnegie Mellon University is running a computer program called the Never Ending Image Learner (NEIL), which searches the web for images and characterises them in order to make some 'common sense' judgments about the world.
The project is supported by the Office of Naval Research and Google Inc. It will present its findings on 4 December at the IEEE International Conference on Computer Vision in Sydney, Australia.
NEIL makes use of recent advances in computer vision that enable computer programs to identify and label objects in images, to characterise scenes and to recognise colours, lighting and materials with a minimum of human supervision. In turn, the data it generates will further enhance the ability of computers to understand the visual world.
NEIL also makes associations between these things to obtain common sense information. Based on text references, it might seem that the colour associated with sheep is black, but NEIL can realise that sheep typically are white.
'Images are the best way to learn visual properties,' said Abhinav Gupta, assistant research professor in Carnegie Mellon’s Robotics Institute. 'Images also include a lot of common sense information about the world. People learn this by themselves and, with NEIL, we hope that computers will do so as well.'
The program has been running since late July and runs on two clusters of computers that include 200 processing cores. It has analysed three million images, identifying 1,500 types of objects in half a million images and 1,200 types of scenes in hundreds of thousands of images. It has made connections to learn 2,500 associations from thousands of instances.
