Phones have been doing this for a LONG time, if you had a oneplus 7 pro (released may 2019) and took it into an extremely dark area, like where your eyes can't see anything at all, and take a 'night mode' photo, it asks you to hold it steady for about 4-5 seconds and is clearly taking a series of photos with long exposure time. Then it stacks them and uses data from the sensor of hand movement jitter to deblur.
The result when done outdoors (like, standing in a dark forest at midnight) will be a photo where you can actually see what's around you, and details show up as if it was a brightly full moon lit night.
That's a bunch of photos from the same camera, not 2 lenses at the same time
I know, I just don't think it's a huge jump from "take multiple photos over a 2s period from one camera and stack them" to "take several photos from discrete physical cameras and stack them". As a subscription of paid app I am not the target market for the person who made this, I agree with another person commenting in here that said they should be working for a phone manufacturer.
I’d argue the opposite. The geometry involved in reconciling photos from two separate lenses with different focal lengths is quite challenging. If you pay close attention you will see some differences in the before/after samples in the site (for example the ear in that kitchen selfie).
The math is surely more complex than aligning same-lens photography even when you consider the change in angle and perspective between two shots from the same lens.
Not really. Whether you have two different lenses or the same lenses with some offset from hand movement you need to accurately model the lens(es). Once you can do that for one lens you can just as easily do it for two - it's just more variables to optimize, not a fundamentally harder problem. Panorama software has supported images from different lenses since forever.
The Amazon Fire Phone did this in 2014. I'm not aware of any prior art in mainstream devices aside from gimmicks like 3D (stereoscopic) photos. I can't remember if the Fire Phone had any relevant gimmicks but it certainly used multiple lenses to produce a single image.
UAV sensor people made a thing that uses 300+ phone size sensors in a single pod, and software workflow to fuse the output together. It's been around quite a few years now.
https://www.google.com/search?client=firefox-b-d&q=ARGUS+pho...
Ah yes, how could I forget the ARGUS-IS, a household staple of the early 21st century. Those were the days...
I keep an extra one next to my panini press, the image quality from the whiteboard notes area on my fridge has never been sharper!
I think it's interesting that the same functionality appeared on iPhone in 2019 as well. Google tells me it was the iPhone 11 and iOS 13.
Night Sight appeared on the Google Pixel camera app in 2018. I think Nokia was doing it first on phones before Microsoft destroyed the remains of their phone business.
The reborn Nokia/HMD was also doing multi-sensor image stacking back in 2019 on the Nokia 9 Pureview.
The related paper (2019): https://google.github.io/night-sight/
Google has some great blogs about their camera technology https://research.google/blog/night-sight-seeing-in-the-dark-...
There is an iOS app that does this called Spectre: https://spectre.cam
From the authors of Halide.
It's different from just stacking exposures, it's a paid app that does progressive animation of exposures for creative effect and says it's "AI-powered" whatever that means.
It's a paid app that I literally have, and all it does is take multiple photos and composite them together. It can either naïvely average them together in-place like a traditional long exposure, or align them using machine learning. I'm actually appalled they use the term "AI" considering who they are but that doesn't matter, it's ML. I don't use any of the animation crap, but the aligned long exposures can be useful for ultrawide takes since that camera is so noisy.
Edit: actually I can't seem to find the setting that aligns them. I seem to remember it being able to do that but maybe it doesn't... which would be disappointing...
> I'm actually appalled they use the term "AI" considering who they are but that doesn't matter, it's ML.
I agree, and I built it!
The app launched in 2019. Ideally I would have said ML, but it felt like AI had more recognition among non-technical users.
If I launched it in 2026, it would 100% say "ML" not only because it's more technically accurate, but to make it abundantly clear it doesn't generate slop.
I have it too but after like 3 times I never used it again, felt like a cheap gimmick. I regret falling for hype. If anything they do good marketing
People are slapping "AI" on anything these days to try to sell more units of whatever it is.