Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

First-Person Hyper-lapse Videos

research.microsoft.com

Post reply on HN
Print view View original

101–110 of 184 posts

Re: First-Person Hyper-lapse Videos

#101

Post by m_mueller »

tibbon wrote: post #98

I'll be the immature one and say I'm curious how funny porn would turn out with this speeding it up by 10x. They don't show much of how it deals with people and I imagine the results would be terribly funny looking- but perhaps awesome. Also, I will pay $$$ for this to use with my motorcycle footage from GoPros.

My guess is it doesn't work well with footage with cuts in it, it's probably meant for continuous shots (there are a few long ones in Hugo and Gravity for example, so this might be interesting input material). From how the climbers appear, porn would probably decompose into static images.

Re: First-Person Hyper-lapse Videos

#102

Post by itchmasterflex »

limsup wrote: post #99
itchmasterflex wrote: post #97

Would it be possible to do something like this for audio? It would be incredible to sample an hour-long album or mix in minutes.

time is kind of an essential component to music...

Time is supposed to be essential to video as well.

This video http://vimeo.com/13669078 has time-lapsed audio. I wonder what it would sound like using the hyperlapse effect.

Re: First-Person Hyper-lapse Videos

#103

Post by iLoch »

doomrobo wrote: post #70
fastball wrote: post #42

Earlier quoted context omitted.

$1 seems cheap.

The processing is quite cheap for a company with its own datacenters and computation clusters. Not so cheap for an individual though. So a user could pay a dollar while Microsoft is only spending pennies.

I think fastball is talking about the value proposition. Capitalism would suggest a higher price point I'd imagine.

Re: First-Person Hyper-lapse Videos

#105

Post by mamoriamohit »

dublinben wrote: post #29
UnoriginalGuy wrote: post #7

The result is quite simply breathtaking. It looks like something shot for a movie using a stabilised dollycam, the fact they were able to achieve the same thing using nothing but a GoPro, their software, and likely a week of post-processing on a high end desktop PC is simply amazing. I hope we see this technology actually become readily available. There might still be work to be done, but in general if they can repro…

It's certainly way better than the source video, but it's nothing close to what would come from a steadycam or dolly. You couldn't use this finished product in any kind of real production.

For hobbyists like me, it is amazing.

Re: First-Person Hyper-lapse Videos

#108

Post by lifeformed »

zodiac wrote: post #91
steven2012 wrote: post #86

Earlier quoted context omitted.

Actually, yes it's exactly what I want, I'm not sure why you would think I misunderstood what the software did. I have hundreds of gigabytes of 1 sec timelapse photos that could convert into pretty neat movies given what I saw.

I think what NeilSmithline is trying to say is, the microsoft technology is designed to convert regular 24fps videos to 2.4fps videos (10x time lapse). If you feed it 1 sec timelapse photos (ie, video at 1fps), the output would be 0.1fps.

Yes, it requires the information from all the discarded frames, in order to reconstruct the 3D world for a smooth curve.

Re: First-Person Hyper-lapse Videos

#110

Post by derengel »

pron wrote: post #52
UnoriginalGuy wrote: post #7

The result is quite simply breathtaking. It looks like something shot for a movie using a stabilised dollycam, the fact they were able to achieve the same thing using nothing but a GoPro, their software, and likely a week of post-processing on a high end desktop PC is simply amazing. I hope we see this technology actually become readily available. There might still be work to be done, but in general if they can repro…

It's striking but it's far more believable when you realize that they need to play at a much faster speed than the source, so they have tons of extra samples from which to extract information. They basically use all that data in the extra frames (that would otherwise simply get tossed away in a regular time-lapse) to construct a 3D scene. This wouldn't look nearly as good if they had to play it at normal speed.

"This wouldn't look nearly as good if they had to play it at normal speed."

For normal speed you wouldn't need this :)

Post reply on HN