Computational photography from selfies to black holes
1–10 of 61 posts
Re: Computational photography from selfies to black holes
#2This is a truly incredible article.
And even a mind-bending (or optic?) glimpse into the future.
Thank you for writing it.
Re: Computational photography from selfies to black holes
#3> This approach pioneered by the guys, who liked to take pictures of star trails in the night sky. Even with a tripod, it was impossible to shot such pictures by opening the shutter once for two hours
Yes, but also you can't leave a digital sensor collecting for two hours, the pixels start saturating and the noise builds up- not like film, which can do truly long exposures.
Re: Computational photography from selfies to black holes
#4This is a great article!
> In fact, that's how Live Photo implemented in iPhones, and HTC had it back in 2013 under a strange name Zoe.
A reference to zoetropes, which were arguably one of the first "movies". https://en.wikipedia.org/wiki/Zoetrope
> To solve the problem, Google announced a different approach to HDR in a Nexus smartphone back to 2013. It was using time stacking.
I don't think time stacking is the appropriate term to use here, as standard HDR is also doing "time stacking", in that it takes multiple photos with different exposures across a small interval of time. Maybe "Fixed exposure fusion"?
I think there's a lot more to be done in computational photography, in research and engineering. In fusion of images from multiple cameras, we're barely scratching the surface. Exciting times ahead!
Re: Computational photography from selfies to black holes
#5On the extreme end, if you live in the countryside you can replace GPS entirely using celestial navigation/star tracking. This is commonly used for rockets and satellites but right now if you own a truck or pickup you can easily go completely off-grid by mounting a lens on the rooftop.
E.g.
http://nova.astrometry.net/
For implementing the system on an embedded device e.g. toy drone, raspberry pi etc., the main data structure you want is a k-d tree together with some sort of evergreen star chart (it doesn't have to be extremely evergreen, current astronomy libraries can easily predict orbits for a couple decades without significant skew/deviation unless you are are aiming for centimeter level geolocation accuracy).
For the hardware you can either use existing consumer-grade stuff followed by a ton of image processing with ML as suggested above or you can use a industrial grade tracker which easily exceeds 4 figures.
https://blog.satsearch.co/2019-11-26-star-trackers-the-cutti...
It's a pretty fun weekend project. Here are some links to get started:
https://github.com/mrhooray/kdtree-rs https://github.com/astronexus/HYG-Database/blob/master/READM...
Instead of jacking up your truck, add celestial nav to it. Nothing screams freedom and independence more than cutting dependency on state-funded satellite systems. Caveats: needs more signal processing during daytime, fallback to inertia navigation when it is cloudy.
Re: Computational photography from selfies to black holes
#6This is all really cool, but, there's one thing you can't make up for enough with processing: optical zoom, (digital zoom, however much temporal super resolution trickery, has a different angle).
Re: Computational photography from selfies to black holes
#7On the extreme end, if you live in the countryside you can replace GPS entirely using celestial navigation/star tracking. This is commonly used for rockets and satellites but right now if you own a truck or pickup you can easily go completely off-grid by mounting a lens on the rooftop. E.g. http://nova.astrometry.net/ For implementing the system on an embedded device e.g. toy drone, raspberry pi etc., the main data s…
Rockets and LEO satellites often use GPS these days because it's easier, but, I am impressed by this diy startracker.
Re: Computational photography from selfies to black holes
#8This is all really cool, but, there's one thing you can't make up for enough with processing: optical zoom, (digital zoom, however much temporal super resolution trickery, has a different angle).
DSLRs have exceptional lenses with exceptional but large sensors, which means the zooms are modest (up to around ~600mm)
So the really interesting long focal lenght cameras are the all-in-one superzoom cameras like the canon sx70hs and nikon p1000.
They accomplish the high magnification by using a sort-of-good lens with a sort-of-good small sensor, achieving up to 3000mm "equivalent" zoom.
Unfortunately, the "pro-sumer" design gives you an electronic viewfinder and slower less accurate focus and all kinds of other non-dslr mediocrity.
sigh.
Re: Computational photography from selfies to black holes
#9So someday we can expect a $1000-ish smartphone which image quality-wise can compete with medium format gears like Fuji or Hasselblad?
Re: Computational photography from selfies to black holes
#10This article is completely, utterly wrong as soon as it starts talking about anything plenoptic. Please disregard after that point as there are serious factual errors, particularly regarding what Google did or didn't do with Lytro.
(Source: I'm in the VR / plenoptic space, knew a bunch of people at Lytro, some of whom are now at Google. Timelines and facts do not match this article's assertions.)