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

Motion Estimation with Optical Flow

blog.nanonets.com

Post reply on HN
Print view View original

1–10 of 24 posts

Re: Motion Estimation with Optical Flow

#2

Post by ole_gooner »

Hey,

Most real-time video processing systems/techniques only address relationships of objects within the same frame, disregarding time information. Optical flow accounts for this temporal relationship between frames. Advances in Optical Flow have changed the game in Object Tracking and Human Activity Recognition in videos.

This article explains the fundamentals and gives you the code to try it out for yourself.

Re: Motion Estimation with Optical Flow

#3

Post by haxiomic »

Back in 2008 there was an amazing demo of realtime dense optical flow on a GPU[0] but all the links are dead now. I've searched hard for another comparable implementation since then but had no luck.

Does anyone have a hint on what technique they might have used?

[0] https://www.youtube.com/watch?v=ssINeWRb58M

Re: Motion Estimation with Optical Flow

#4

Post by Darkphibre »

I've dabbled with optical flow for a hobby side project (using the Windows ShaderEffectClass of all things). I worked on the Kinect back in the day; while I was primarily audio, I had a fascination of applying DSPs to two- and three-dimensional temporal data.

I've always felt that it was a missed opportunity to tap into temporal information for entity recognition. I'm excited to see this take hold!

Re: Motion Estimation with Optical Flow

#5

Post by pontifier »

I hit upon a similar idea back in about 2000... the key at the time was video encoding hardware that was designed to incorporate MPEG compression. Pulling motion vectors from the encoded stream could have given realtime optical flow way back then. Unfortunately I was nowhere good enough as a programmer to take advantage of it for real projects. A team at MIT had a paper about this that I was able to find.

Had a page about it on robots.net, but it seems to be down now

Re: Motion Estimation with Optical Flow

#6

Post by tbirdz »

pontifier wrote: post #5

I hit upon a similar idea back in about 2000... the key at the time was video encoding hardware that was designed to incorporate MPEG compression. Pulling motion vectors from the encoded stream could have given realtime optical flow way back then. Unfortunately I was nowhere good enough as a programmer to take advantage of it for real projects. A team at MIT had a paper about this that I was able to find. Had a page…

We're doing this with H.264 at work. Pretty useful since we need to generate the encoded stream anyway, so now we get motion estimation for free.

Re: Motion Estimation with Optical Flow

#7

Post by lcrs »

haxiomic wrote: post #3

Back in 2008 there was an amazing demo of realtime dense optical flow on a GPU[0] but all the links are dead now. I've searched hard for another comparable implementation since then but had no luck. Does anyone have a hint on what technique they might have used? [0] https://www.youtube.com/watch?v=ssINeWRb58M

A lot of that group's publications are listed here, many involving optical flow: http://web.archive.org/web/20161014025823/http://gpu4vision....

Maybe this one from 2007: https://www-pequan.lip6.fr/~bereziat/cours/master/vision/pap...

Re: Motion Estimation with Optical Flow

#8

Post by haxiomic »

lcrs wrote: post #7
haxiomic wrote: post #3

Back in 2008 there was an amazing demo of realtime dense optical flow on a GPU[0] but all the links are dead now. I've searched hard for another comparable implementation since then but had no luck. Does anyone have a hint on what technique they might have used? [0] https://www.youtube.com/watch?v=ssINeWRb58M

A lot of that group's publications are listed here, many involving optical flow: http://web.archive.org/web/20161014025823/http://gpu4vision.... Maybe this one from 2007: https://www-pequan.lip6.fr/~bereziat/cours/master/vision/pap...

Oh amazing find Icrs, thank you! :)

Re: Motion Estimation with Optical Flow

#10

Post by rjeli »

tbirdz wrote: post #6
pontifier wrote: post #5

I hit upon a similar idea back in about 2000... the key at the time was video encoding hardware that was designed to incorporate MPEG compression. Pulling motion vectors from the encoded stream could have given realtime optical flow way back then. Unfortunately I was nowhere good enough as a programmer to take advantage of it for real projects. A team at MIT had a paper about this that I was able to find. Had a page…

We're doing this with H.264 at work. Pretty useful since we need to generate the encoded stream anyway, so now we get motion estimation for free.

Awesome. Which encoder provides this? Can you pull it from cuvid?
Post reply on HN