The Semantic Robotic Vision Challenge

A robot challenge has been announced that will test robots' abilities to recognize objects as well as their skill at understanding language.

The robot contestants will be given a list of objects located around their competition arena where they will be expected to go get them.

The robots will be required to look up the name of the objects on the internet - for example from Google image search - then find the required items.

From the website:

Integrating a mobile robot with the vision research adds another interesting layer of complexity that would not ordinarily be available in a purely computer vision competition. Semantic understanding of the objects could also be expanded to scene understanding as well.

So the objects will be located in proper context for example office items will be set up in a simulated office.

The final rules for the competition will be posted this week. Contestants are required to pass a qualifying test to get into the finals.
The final event will be held at the Twenty-Second AAAI Conference on Artificial Intelligence held in Vancouver July 22-26 2007.

The Semantic Robotic Vision Challenge