[ExI] Google AI low-resolution photo enhancemen

Dave Sill sparge at gmail.com
Fri Sep 10 20:47:22 UTC 2021

On Fri, Sep 10, 2021 at 3:47 PM Ben Zaiboc via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On 10/09/2021 05:24, BillK wrote:
> > I think the AI does backward and forwards resolving to check that the
> > final photo does blur back down to the original photo, to check that
> > the expansion is valid
> I dunno, that just sounds wrong to me. Surely more than one image can be
> blurred down to exactly the same pattern? Extrapolating then simplifying
> again doesn't seem like a good check on the validity of the extrapolation.

Exactly. To go from low-res to high-res without additional information
requires making up new data. The Google photo enhancement AI is good at
making up data that results in believable portraits--and it only does
portraits at this point.

None of the stories about this enhancement have compared the results of
enhancing an image with the actual high-res version of the portrait. They
all published the same Google-supplied images which were no doubt
cherry-picked to make the software look good.

Again, this technology will be fine for upscaling images for non
law-enforcement purposes, but it in no way recovers data lost due to low

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20210910/2a7e5a3f/attachment.htm>

More information about the extropy-chat mailing list