On 2 May 2010, at 12:16, Randall Lee Reetz wrote:
Are you kidding?
No.
These are revisionist applications exactly as I have noted.
Again, I have no idea what you are talking about.
The only revolution in photo programs in the last 20 years is face recognition.
Why is facial recognition revolutionary, but content-aware fill, non- destructive/parametric adjustments etc. etc. *not* revolutionary?
But what has been done with it? Almost nothing.
What's 'almost nothing'? There are apps such as iPhoto which can auto- crop images to show just the faces, cameras which use facial recognition to choose where to focus or wait for the subject to smile to take the photo. At the moment you are just making yourself look ignorant.
Is it available to rev programmers?
As far as I know there are command-line apps around.
Can it be generalized to learn any object?
It's the other way around - facial recognition algorithms are mostly based on more general feature-recognition code (feature as in distinct areas of an image, not necessarily part of a face). Look up SIFT and similar stuff, it's what's revolutionised all kinds of panorama stitching and image alignment apps over the last 5-6 years.
Does it get better as we all work with it?
Yes, most software that includes facial recognition *only* works at a useable level by using user-confirmed results to improve the hit-rate.
Ian _______________________________________________ use-revolution mailing list [email protected] Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
