I’m very very interested in this research at MIT which is basically tackling vision and the related learning issues.
I’ve started on the paper provided at the top of this webpage entitled “A Theory of Object Recognition: Computations and Circuits in the Feedforward Path of the Ventral Stream in Primate Visual Cortex” which is a comprehensive description of the model. I’m about 15 pages in and I’m quite impressed.
The first 2 pieces of software are definitely of interest, especially the 2nd one, hmin which will be good for starters. It is in c++ and matlab which are standard for vision, however, since our LIDA framework is in Java, I am looking into a port or a Java version.