Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by bitL on Wednesday October 18, 2017 @10:01AM
(#15499185)
I wish my Nikon pro camera had a similar chip inside. Now we have two worlds - awesome pro cameras with incapable post-processing chips (I mean beyond RAW and initial processing) and horrible mobile cameras with outstanding post-processing chips.
[ Parent
| on HN ]
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by ggus on Wednesday October 18, 2017 @10:29AM
(#15499422)
Also the double lens, one for wide angle, one for tele, snapping at the same time and combining the images into one.
It will be a looooong time until we see that on a pro camera with interchangeable lenses, if ever.
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by ksk on Wednesday October 18, 2017 @12:46PM
(#15500656)
We already have zoom lenses that can do wide angle and take portrait shots. Having multiple full frame sensors with multiple lenses doesn't make much sense to me TBH. What is the problem you're trying to solve here?
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by ksk on Wednesday October 18, 2017 @12:49PM
(#15500686)
Sony tried to do it with their alpha series of cameras. You could install apps from sony, and do post-processing. The cameras were under-powered for this sort of stuff, and the experience of installing those apps isn't very seamless.
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by bradleyjg on Wednesday October 18, 2017 @12:32PM
(#15500539)
If you have burst mode and can get the raw frames, then shouldn't you be able to recreate the effect in post-processing? With the severe power constraints of a mobile phone a co-processor makes sense, but I'd think a general purpose desktop processor could easily handle this algorithm.
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by londons_explore on Wednesday October 18, 2017 @01:12PM
(#15500894)
Can you burst 80 frames @ 60fps of 12 bit 12 megapixel RAW and save all that output, together with the output of gyroscopes and accelerometers timed to the start and end of exposure of every row of the image?
If you can, you might be able to postprocess it... But as far as I know, no DSLR cameras even have gyros, nor sufficiently high burst rates, so that kind of photography is out of the question.
It comes out to over a gigabyte of data per image... and you can't compress it before processing...
Re:Pixel Visual Core: Google’s first custom-designed co-proc...
by bradleyjg on Wednesday October 18, 2017 @09:54PM
(#15504894)
Good point about the gyros and accelerometers. I'd think you could do decent lossless compression given that the frames should very similar. I suppose though that once you are doing sophisticated compression you aren't that far off from implementing the post processing we are talking about.
I guess the long and short of it is these camera companies aren't software people--except isn't Sony a big camera company? You'd think they'd be able to bridge the gap.