At the time of this writing, we have a working hardware test platform based on the Sony Vaio Picturebook, as described above. We have a functioning classifier system, with preliminary training results of 80% correct acceptance to 24% false acceptance in the task of identifying six events after a relatively short training period.
We have a working agents-based middle-wear framework and preliminary Memory Glasses application, which is not yet integrated with the functioning classifier system. The primary task to accomplish before we can start full-scale testing of the system is integrating the functioning classifier system with the middle-wear and application layer.