By 
Early this afternoon, the world became enamored with Instagram’s latest mobile app since launching Bolt a few weeks ago. It’s called Hyperlapse and the app essentially allows users to create beautiful time lapse videos using nothing but their smartphones. Sure, there are a few apps that already do something similar, but it’s nothing close to the way Hyperlapse handles time lapse video.
Instead of taking photos and later converting them into a video, Hyperlapse simply records video (doesn’t snap photos), showing the real-world time passed and time of the outputted sped up video. Once finished, Hyperlapse is able to stitch together snaps shots into a single, incredibly stabilized video. Really, it’s pretty remarkable and results speak for themselves.
Hyperlapse video in action (click to view)
The only problem is the app, like many others, is only available on iOS, leaving Android users out in the cold. Because we’re sure, like us, many of you are wondering when an Android version will be available you probably shouldn’t hold your breath. Before you get upset, this has less to do with any sort of exclusivity deals or platform favoritism.
In a statement given to Wired earlier today, Instagram says Google will have to make some changes to the camera and gyroscope APIs on Android before Hyperlapse could ever become compatible with our devices. Bummer. The good news (if any) is that Instagram is ready and willing, so the future of Hyperlapse for Android isn’t entirely bleak.
For more info on Hyperlapse, check out the trailer down below and if you want to try it on your iPad, you can find it on iTunes here.
I find that excuse hard to believe. Plenty of other 3rd party apps can access the camera and gyroscope. What makes this so special?
I agree. I think Instagram is just trying to apply their iOS code to Android. I hate when devs do that.
apparently you know more than the developers?
I never claimed to.
Admittedly it looks like my last comment was scrubbed / removed for perhaps being a bit too emphatic in my response. I’ve been working with next-gen video engine dev on Android since it has been possible with Android 4.3. I’m moving toward launching next-gen video engine tech for Android including timelapse / stabilization tech implemented in the same way that Hyperlapse is implemented. I’ll be out at the Big Android BBQ, so look forward to demoing this tech to Phandroid folks on hand and Android enthusiasts at large. As Barry mentions in another comment… This is an excuse more or less as this tech is possible on Android. There is going to be some fantastic media apps available on Android soon.
Sounds like a BS excuse that people are very ready to believe because of all the nonsense Apple tries to spread about Android as they slowly copy it.
We’ll see if it’s BS if clones start popping up in the Play Store.
can someone say “exclusivity deal”….
Didn’t Microsoft come up with this, or something like it, a while ago?
And there they go again.. making a complementary product, to fuel my already insatiable Instagram addiction :)
Have the fixed the issue with overly compressed image uploads in the android version yet? If not then I wonder how much any hyperlapse would be screwed even if they did include that feature for Android.
That’s some next level BS right there
My Note 2 and my wife’s MotoX already can do this with the stock camera/video app.
Hyperlapse looks like a great effect to have fun with. It’s somewhat believable the way they implemented this might not work with Android APIs, however I’m sure there is some other way to do it.
It would be great if Google just shrugged and built it into Android L’s camera app.
Google really needs to put some better effort in their camera software, but after 4 years I’m still guessing they won’t
I don’t actually know how Hyperlapse is different than animated stills or frame-dropped videos (fast motion). Can someone explain, please?
Edit: I get it: http://www.wired.com/2014/08/hyperlapse-instagrams-new-app-is-like-a-15000-video-setup-in-your-hand/
It is about image stabilization. That’s it. These videos can be run at normal speed as well and they appear to have been made with a steadycam. What makes them different from animated stills or sped-up video is that the jitter correction is done (live?) based on movement the gyroscope senses, not based on a review of the video done frame-by-frame on a more substantial computer. This difference enables your phone or tablet to produce these. A faster PC is not needed.
In the end, though, you get nearly the same result. Frames are adjusted and cropped to remove the shaking of the camera.
That’s why camera software for some OEM’s, like Samsung for example, have image stabilization built into it.
The first developer(s) to make it into an app for android will be a billionaire though.
And yes, I’d be willing to bet money that apple made back door dealings to do this, they have done it before and will do it again. But with Android having over 80% of the world market I’m sure Instagram will have this with in a year on Android, and won’t let someone else beat them to it like Vine or someone else.
Great, Apple paid them to make it an exclusive AND give a BS excuse that talks down Android.
Cant you just come to the terms that this is an Android issue? Why do you need to blame apple for it?
Will apple buy instagram soon or something? I’m thinking of uninstalling it until they come up with a better excuse!
instagram is already owned by Facebook, so, no, don’t expect a purchace from Apple.
Here’s a better excuse. Android’s developer tools are not on par with iOS.
What else is new.
So, I’ve had Karpenko’s research paper (the fellow referenced in the Wired article / involved with Instagram) on the gyroscope approach bookmarked for well over 2 years at this point. *I’ll post a link in the first reply* What it seems that the gripe and misdirection from Instagram’s statement about Android is that the coefficients necessary to create a universal solution ecosystem wide on Android requires more complicated solutions such as a device database or a potential configuration step to determine the correct values for any given device. An example of some of the important parameters for the calculation are the focal length of the camera, the potential delay between gyroscope readings that may differ between different sensor hardware manufacturers, and just a handful of other parameters. Even in the paper the technique necessary as part of a configuration step is fully described and is straight forward. It only needs to be accurately performed once per device (or potentially similar hardware components across devices) then loaded into a device database really.
The reason these steps aren’t necessary on iOS is because there are only ~6 devices or so that need to have these coefficients determined and they can be hard coded into the app thereby not requiring end users to need to configure them manually.
To my knowledge _neither_ Android or the iOS APIs / SDK support querying the coefficients necessary for the configuration of this particular technique. It’s just that one can experimentally derive them and hard code them for a handful of iOS devices fairly easily.
The sad thing if what I wrote above is the actual reason is that it ultimately amounts to _extreme laziness_ on the part of Instagram / Facebook, a company with infinite funding & resources, such that it is easier to put out a statement of misdirection against Android than actually tackle what is a solvable problem and one that I as a single independent bootstrapped developer will be doing with my hyperlapse implementation for Android. QED
Karpenko’s and others research paper:
http://hci.stanford.edu/cstr/reports/2011-03.pdf
A recent paper from NVidia research on the topic w/ more details:
https://research.nvidia.com/publication/non-linear-filter-gyroscope-based-video-stabilization
Android here we come! ;)