A Way to Bring Computational Photography to Your Camera?

Biro

Hall of Famer
Location
Jersey Shore
Name
Steve
An ad for this device popped up on Facebook today. I'd never seen it before. I'm trying to decide if this really is a way to bring computational photography to our dedicated cameras, or whether it's just an extra Intelligent Auto algorithm on top of what our own camera might (or might not) offer. Has anyone else seen this or know anything about it?

 
An ad for this device popped up on Facebook today. I'd never seen it before. I'm trying to decide if this really is a way to bring computational photography to our dedicated cameras, or whether it's just an extra Intelligent Auto algorithm on top of what our own camera might (or might not) offer. Has anyone else seen this or know anything about it?

It looks like an interesting idea but two things struck me. First, even though they say it's more than just a filter the end results on their before and after images sort of look like filters. To me, they also look overprocessed and if you like the look that's probably fine but if you don't (and I don't) it's not clear if there's a way to adjust it. I imagine the output is determined but the choice of images used to train it. And training is the second point - they say they train it on "thousands" of photographs, but that's not very many. Most of us probably have thousands of photos on our drives. Adobe's recent post about their super resolution machine learning feature says it was trained on "millions" of photographs just to achieve one objective.
 
I am using the Topaz suite in my workflow, when needed. Interesting pieces of software, they are. The major task at hand for those things is for me DNG converting, upsizing and some other stuff from the early onset days of a 3MP P&S, and some diverse tasks in what is a major task, digitalizing about 120 years worth of family photos. I keep pushing on the start-up time of the latter one.

Auto removal of dust and gore would have been very nice.
 
Back
Top