ROD4 CHOP

[bLackburst]

Hi Derivative

I’m currently trying to source a demo model of the ROD4 scanner for testing with Touch designer and was hoping you guys could give me a bit more information on its compatibility and use with TD. Any and all help appreciated.

Which models are supported?

What is the maximum number of blobs produced?

How much work needs to be done to the incoming data for it to make sense spatially? I have heard that the chop doesn’t actually output blobs, is this correct?

What is the object detection mode (not supported currently, I know) and are there plans to continue development on the ROD4’s support?

Do you have a basic working example you can provide or perhaps a best-practice document for calibration procedure etc?

Thanks.

[malcolm]

Hey,

Any of the ROD4 models should work I think. We've only tested with on which was the 50Plus if I remember correctly. There is no limit to the number of blobs that can be detected.

Yes, it outputs blobs, or the measured distance is the alternate mode.

We no longer have one of the devices in the office so no future development is currently planned. Future development could be done using Pro support hours and loaning us the device you wish to use.

Sorry, we don't have any example file for this particular node right now.

[matthewwachter]

We use the ROD4-50Plus

I have a nice little component that makes it pretty easy to calibrate. I just have a bit more work to do on it before posting. Hopefully I'll have it posted some time early next week.

[bLackburst]

Thanks for the replies.
Could anyone tell me if the ROD4 is capable of broadcasting data to a concurrent backup machine?
I'm also looking at radar touch and OptiTUIO. Any experience with these products?

[Ennui]

Thanks for the replies.
Could anyone tell me if the ROD4 is capable of broadcasting data to a concurrent backup machine?
I'm also looking at radar touch and OptiTUIO. Any experience with these products?

We've tested RadarTOUCH and it worked really well for us. Had it setup with touch and working fine.

[bLackburst]

After playing with the ROD4 a little, I'm a little disappointed in how noisey the blob readouts are and how sensitive to false positives and false negatives. For instance, for use as a multitouch display, I'm guessing nobody else is able to resolve 2 fingers on the same hand, or even hands closer together than 25cm or so without the blobs jumping around, increasing id's etc.
I'm sure you could spend a week programming around a couple of things, but I'm a bit surprised at how satisfied everybody seems to be with them.
Has anyone used "Radar Stitch", the radarTouch application that stitches multiple sensors together? I'm just getting a bit sick of looking at videos that are obviously well-selected takes of best-case scenarios and need some real opinions, advice, vids, shared components etc.
Thanks again.

[CutMod]

When I was working with ROD4 on a dancer tracking project, I found the built-in blob mode to be troublesome and instead used the raw sensor data. That was over the summer and I haven't had a chance to revisit since...

Given that the thing has like 60m range, I'm not sure how good of small object tracking (i.e. fingers) you should expect. I remember reading in the ROD4 manual that accuracy has sharp falloff as distance from sensor increases. Not sure where optimal range for finger differentiation would be...

[bLackburst]

The airscan and radar touch implementations have a couple more tweaks like predictive algorithms according to the docs

[ziopix]

We use the ROD4-50Plus

I have a nice little component that makes it pretty easy to calibrate. I just have a bit more work to do on it before posting. Hopefully I'll have it posted some time early next week.

Hi Matt,
I would really appreciate if you can share your component and experiences about the ROD4. I'm not really understanding how it's working in Raw Data mode and in blob tracking mode it doesn't seems really reliable, I mean after 5-6 time it detect the blob it stop working.

Any suggestions are appreciated,

Thanks