Who Needs A Kinect? Eye Synth OSC Lets You Make Music With Your Web Cam For $.99

Can’t afford to drop $150 on a Kinect so that you can experiment with gestural music control?

How about $.99?

Dustin O’Connor’s Eye Synth OSC is a $.99 Mac app, available in the App Store, that lets you use an i-sight or web camera to make music based on motion.

Eye Synth OSC splits your camera up into 8 individual sections. You specify which section corresponds to the notes you want. It comes with a built-in synth.

Eye Synth OSC also transmits the notes, via Open Sound Control protocol, so that you can use Eye Synth with any music application that supports OSC.

Check out the video demo for an example of Eye Synth OSC in action.

This looks like it shares the limitations of low resolution and higher latency of the Kinect. But if it doesn’t do what you need, at least you’re only out $.99.

If you’ve tried Eye Synth OSC, share your thoughts in the comments!

36 thoughts on “Who Needs A Kinect? Eye Synth OSC Lets You Make Music With Your Web Cam For $.99

  1. I love it that this helps show that the kinect really isn't all that special. Mo-cap has been around for many years… selling it as a consumer device doesn't suddenly create unheard of new uses for it.

  2. I have installed this and my initial problem with the app is that there is not visual feedback on where the 8 sections are and when a section is triggered. So I feel like I am just waving my arms around in the vain hope of acheiving something.

    Documentations with diagrams to somehow indicate how this works are also thin on the ground. But hopefully the developer will listen to feedback and update it.

  3. More options are good – even if they are not perfect. Electronic musicians shouldn't be limited to a computer keyboard or a synth keyboard.

  4. Mo-cap has been around for many years, but mo-cap with an independent "depth" channel has not. Moving from 2 dimensional tracking to 3 dimensional tracking gives you a lot more options for performance variables…

  5. Agreed. Guy waves his hands around as if he's grooving to the bad music. He needs to make a better attempt at marketing; from this video we've learned nothing about the app at all. Could have potential tho!

  6. Agreed. Anyone who things you can wave your arms around and get better/faster results than just typing on a keyboard is being dumb.

  7. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  8. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  9. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  10. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  11. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  12. What? Motion capture has always been in three dimensions. That was the whole point of it… to capture what real humans do and map it to three dimensional geometry which can be rendered as two dimensional images. It's been used in movies for almost two decades, and in sports games for at least 10.

    Granted, having two cameras in the kinect instead of three like a pro-level system has is "different", but the resolution goes down, not up. My original point was that if there was some overwhelmingly awesome use for consumer based motion capture, it would almost assuredly have been discovered over the last couple decades.

    Check out this video of a guy demoing a minority report style motion control system. He shows some 15 year old ideas and tech (which are mostly bad). These guys were invited to design the Minority Report interface, and then built it as real tech. 15 years, and nobody has found a good use for it.
    http://www.ted.com/talks/lang/eng/john_underkoffl

    Jump to about 14:00 where he talks about how common this stuff can be. Then go to 14:40 where he is asked, "What is the killer app"?. He doesn't have one. It's got some "wow" factor, but this style of UI is actually pretty limited, not better than what we already have.

    Then go here and check this out. It's a talk about how the cg faces were done for the Benjamin Button movie. Jump to 8:00 if you want to get right to the point. Mocap wasn't accurate enough (even in 3D), so they moved to a 2D capture method that tracked facial contours. This represents the leading edge of this field of tech… the kinect does not.
    http://www.ted.com/talks/ed_ulbrich_shows_how_ben

    A simple google search turns up boatloads of mocap info. It's not new, and it's not amazing. Microsoft has just made a budget bin version of it and put in in your living room. The tech is extremely limited, so using other limited tech (like web cameras, as this poster did) to do get similar results is not a reach.

  13. You raise valid points.

    What makes the Kinect, GMS & this interesting, though, is that they are mainstreaming gestural technologies.

    My interest is not that this is cutting edge, but that it's becoming technology that's accessible for even "trivial" uses.

Leave a Reply