Thursday, March 10, 2011

Doing good with computer vision

Human vision is a hugely complex, and poorly understood, capability, and only those of us who work with machine and computer vision appreciate the gulf between what our eyes can do and what our cameras are capable of. Researchers are of course eager to mimic human perception, but progress is incremental.

One of the organizations that seems to be making headway is the Smith-Kettlewell Eye Research Institute in San Francisco, (S-K for short,) and they seem to be the only organization attempting to combine physiological research with computer vision.

Under the heading, “Applications of Computer Vision for the Blind and Visually Impaired” you’ll find some interesting projects. These use computer vision in a number of ways to assist the visually-impaired. What’s particularly notable is that S-K have recognized most of us now carry around our own imaging system – our cell phone – and all that’s needed is a way for the phone to communicate image information to the user.

I say “all that’s needed…” but clearly there’s a lot of work to be done and I suspect that S-K, like most organizations are resource-limited. Perhaps, given how this work could make such a difference to the lives of many, some of the leading machine companies would be interested in getting involved.

Well it’s just a thought.

No comments: