Will your microphone be a bag of potato chips someday? Maybe….

[Walter Soyka]

Great! Now I understand why we need 8K video!

Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive   |   RenderBreak [blog]   |   Profile [LinkedIn]

[Dennis Radeke]

Adobe was involved in this research as well. I posted the article link to my twitter feed yesterday

https://twitter.com/dradadobe/status/496402301578674177

Extracting audio from video only. Fascinating… http://t.co/RxwE8i32hk

— Dennis Radeke (@DennisRadeke) August 4, 2014

[Jeremy Garchow]

So, is it safe to say that the technology is all that, AND a bag of chips?

[Ronny Courtens]

So you need a 80,000 dollar high-speed cam to be able to recover audio that you could simply have recorded with a cheap microphone? Now that’s what I call progress!

– Ronny

[Bret Williams]

New meaning to we’ll fix it in post.

[Walter Soyka]

All kidding aside, it’s cool research.

I wonder how much spatial and temporal resolution you’d need to measure the differences in vibrations across objects in different locations in space to get visual sonar-assisted roto or keying.

Walter Soyka
Designer & Mad Scientist at Keen Live [link]
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
@keenlive   |   RenderBreak [blog]   |   Profile [LinkedIn]

[Jason Porthouse]

Interesting ramifications for the surveillance state there – high speed high-res cameras able to convert any object into a microphone… *gets tinfoil hat*

Seriously bloody clever though.

_________________________________

Before you criticise a man, walk a mile in his shoes.
Then when you do criticise him, you’ll be a mile away. And have his shoes.

[Bret Williams]

Coming to an episode of CSI in the near future no doubt. After they first zoom in and “enhance” the image to see how many freckles a particular person in a crowd has they’ll convert the visual sound waves vibrating off their nose hairs to see if it matches a voiceprint from an enhanced background conversation over a cell phone.

[Bill Davis]

[Bret Williams] “they’ll convert the visual sound waves vibrating off their nose hairs to see if it matches a voiceprint from an enhanced background conversation over a cell phone.”

New audio technical term ahead: Signal to nose ratio.

Know someone who teaches video editing in elementary school, high school or college? Tell them to check out http://www.StartEditingNow.com – video editing curriculum complete with licensed practice content.

[Mark Suszko]

“I wonder how much spatial and temporal resolution you’d need to measure the differences in vibrations across objects in different locations in space to get visual sonar-assisted roto or keying.”

-Walter Soyka

I think we’ll be seeing something like that pretty soon,based on the same technology used in the Lytro variable-post-focus still cameras, and/or IR beams picked up by the camera lens and the same sensor that images the visible light.

You won’t need green or blue screens any more: just a foot or so of separation distance between discrete elements in your composition, and you will be able to key out independent elements based on their distance from the screen. Eventually, it will also work in realtime, not just post. The computational power to achieve this is rapidly approaching.