Answering Niners Questions about Jelly Bean—The Kinect Drivable Lounge Chair
- Posted: May 06, 2011 at 9:40 AM
- 37,293 Views
- 10 Comments
Download
How do I download the videos?
- To download, right click the file type you would like and pick “Save target as…” or “Save link as…”
Why should I download videos from Channel9?
- It's an easy way to save the videos you like locally.
- You can save the videos in order to watch them offline.
- If all you want is to hear the audio, you can download the MP3!
Which version should I choose?
- If you want to view the video on your PC, Xbox or Media Center, download the High Quality WMV file (this is the highest quality version we have available).
- If you'd like a lower bitrate version, to reduce the download time or cost, then choose the Medium Quality WMV file.
- If you have a Zune, WP7, iPhone, iPad, or iPod device, choose the low or medium MP4 file.
- If you just want to hear the audio of the video, choose the MP3 file.
Right click “Save as…”
- High Quality WMV (PC, Xbox, MCE)
- MP3 (Audio only)
- Mid Quality WMV (Lo-band, Mobile)
- High Quality MP4 (iPad, PC)
- MP4 (iPod, Zune HD)
In this video, Clint and Dan answer your questions about Jelly Bean—The Kinect Drivable Lounge Chair. They cover a number of topics, including how they decided on the current control scheme, how much calibration was required, and how much code they wrote on top of the SDK.
Comments Closed
Comments have been closed since this content was published more than 30 days ago, but if you'd like to continue the conversation,
please create a new thread in our Forums,
or
Contact Us and let us know.
Follow the Discussion
I'd like to see how well it controls now that Clint has (I'm assuming) mounted the wheels the right way.
@dentaku: Clint let me drive the base (sans chair) around the hallways here after MIX using an xbox controller since the chair wasn't hooked back up yet and I can confirm that with the wheels on correctly, it does indeed "turn on a dime".
@dentaku: We did film some stuff. The big difference between what was seen at Mix and now is I can do zero turning radius turning. Let me see if I can't get some videos of it turning now
I never thought about the IR interference issues until now. However, this is just the beginning of a long line of Kinect products that as mentioned during the Kinect panel, will hopefully become smaller and more ubiquitous over time.
awsome stuff, thanks Tina, Dan and Clint
The framerate was kinda strange because it didnt show up in the jellybean app. It seems a little strange that you code was to blame @Dan, after all it didnt to alot of stuff. All i can think of is that it sets the ImageSource over and over, but still.. what ever it was i dont think it'll show up in the beta
About sensitivity i actually ment the control sensitivity, if you did alot of tweaking on how much you had to move your hands to move jellybean or if you had to do any smoothing to prevent jittering. you did talk about that a little earlier in the video though
Perhaps somewhat lost in the uber coolness of Jellybean is how quickly Dan and Clint went from idea to "finished" product. Impressive work all around.
Charles
@aL_: As Charles pointed out, we didn't have a lot of time to really add in any smoothing. The motor controller does have the ability to add in accelleratoin curves and I could have added that in as well. If you saw the scott hanselman video from ch9 live when I was driving it, I can drive it with Kinect smoothly with very little issue. Scott on the other hand ...
But I think the jumpyness you're talking about is at about 8:25 into the video, this has to do with how we grab what is deemed the "closest". What I found was most people would move one hand forward on a Z index on top of the X/Y index which caused the hand tracking jitter issue. Right now I believe it says "can i haz the closest plus all their friends within a tolerance" which is what causes one hand to disappear when you move your hands if you don't move them as if your a mime stuck in a box. Since I was the really the only one that was going to drive it, I just had to remember how to keep it tracking my hands.
@Clint:
Very interesting, i can imagine the hand tilt would be a problem as you say. how deep was the region you considered active? 1 inch or thereabouts?
@aL_: what we showed at mix was literally closest point. There was no minimum distance. I plan to add that in before we release the code.
It was all trade offs of time, a lot of decisions became "is this a need or a want" feature? The issue above was something I could demo around so it was a lower priority. Not saying it isn't important, just not as important as making the robot actually move
cool stuff all the same
i dont mean to say that you needed smoothing, it looked fine to me, im just trying to estimate the kind of sensor performance you where getting
Remove this comment
Remove this thread
close