Large conference room

[rodders]

We have a large VTC room (30 or so people) and any distant end connection normally can see the whole room. This means that they have difficulty distinguishing who is actually talking. We do have preset cameras to zoom closer on an area but they are manualy controlled and due to the layout , quite difficult to use.
My question is: Does anyone have any thoughts on this idea.
Each person has a laptop in front of them, how about putting a camera on each laptop and have it activate when its particular user is talking? Could it be done automatically and how would I mix the video/audio into the codec?
Also if it can be done what would the approximate costs be?

[tom9933]

Rodders,
There are a couple of ways to do this. One we talked about many years ago was to make each laptop an endpoint and host a large multipoint call. We scrapped the idea due to the large cost and the various audio issues, but it still may be a good idea in the future. To do it today I would recommend looking at using a control system with a video switcher. The cost is going to vary greatly based on the number of cameras and the complexity of the programming.
I think a better solution would be to automate the control of your existing cameras. It’s pretty easy to do this if you have an automatic mixer with triggered outputs and a microphone near each person. Typically we do this for student shots in our dl classrooms and we generally take a shot of two or three students. Individual shots are nice, but you don’t want to make them too tight because of movement in the seat. Also be careful that you don’t have too much movement because it can be very distractive if you’re on the far side of the call.
To go beyond this I’d recommend either talking with an integrator that does AMX/Crestron programming or looking into one of the Vaddio ControlView boxes. You can find them at: http://www.vaddio.com/products.asp
BTW if you’d like to see what this looks like from the far end I can probably arrange an IP call with one of our classrooms.

[rodders]

Thank you Tom,

That clarified and re-enforced some of the ideas that I had. Now I know that the presets that zoom to 3 people has been set, like you say, to stop close up movement and not to make life easier for the camera operator. I will look into the voice activated preset. I vaugely remember a tandberg unit supposedly having this capability but it did not work so well because of the microphones etc. No doubt things have improved a lot on the Audio selection front.
Thank you for the offer to view your set up but alas I have no external IP capability.

[Shawn Jones]

The Tandbergs do do this when you are using the various mic inputs on the codecs. Essentially, you set up a camera preset so that when the codec senses audio on one mic that is of higher volume than another, it activates that cam preset, and that person, or persons are viewed by the far-end.