[thedrake]

This will help to solve debates like the one about the moon landing. Use this video (https://www.nasa.gov/multimedia/hd/apollo11_hdpage.html - the highest HD I could find) of the flag moving and extract sound. Then extract the sound (if there is any) http://people.csail.mit.edu/mrub/vidmag/ is the site that he mentions at the end or the ted talk given that goes into more about the workflow.

Not a programmer so not sure how to use the code to implement it. Any takers?

[Sir_Cmpwn]

This is going to do nothing to solve the "debate". Moon landing deniers will find an excuse to dismiss this evidence like they have for the rest.

[jwilk]

The conspirators purposefully made the flag from nylon, rather than candy wrappers, in order to foil future attempts of sound extraction.

[333c]

Am I missing something? Wouldn't there be no sound to extract because the moon has no atmosphere?

Or maybe that's your point — if it were faked, there would be sound, otherwise not.

[mikejmoffitt]

As the original footage was done on film, the time difference between the top and bottom of a single frame is zero, unlike a "rolling shutter" progressively scanned sensor. The framerate is also much lower than the 60fps phone camera used in the last example from the MIT video, which was only barely enough to get meaningful data out.

So unfortunately, we probably won't be using this to get much out of old traditionally archived footage - not because of resolution of the frame, but time resolution.