MIT News: MIT, Microsoft and Adobe team presents SIGRAPH 2014 paper describing a way to extract sound from an object micro-vibrations in the video stream: "The Visual Microphone: Passive Recovery of Sound from Video" by Abe Davis, Michael Rubinstein, Neal Wadhwa, Gautham Mysore, Frédo Durand, and William Freeman. While similar techniques based on laser and high speed cameras has been known for a long time, the MIT team manages to get the sound using a regular 60fps DSLR video stream. A Youtube video explains their approach:
Thanks to TG for the link!
Pretty cool. I can't wait for Abby Sciuto to use this on "NCIS" to recover unmiked terrorist conversations from the vibrations of leaves capture by a traffic surveillance cameras.
ReplyDeleteMore and more applications for image sensors. Still more to come I believe. Great thing for our community
ReplyDelete