Re: [rp-ml] low-res scanning - a follow up

Ian,

I don't think you will be able to do this while maintaining a 1M byte limit (reasons are a bit technical), but you could possibly do it at 10M/part. Less than this amount of information would require you sacrifice a lot of detail and resolution. It might help to see an example of what it is you want to model. Are these pretty much convex items or do they also have a lot of convex regions or holes? Even at 10M/piece, you could still store 1000 on a DL DVD or BlueRay disk and, of coarse, it wouldn't be any problem on a HD, even if each required more than 10M.

G.S.

________________________________
From: Ian Gibson <mpegi@nus.edu.sg>
To: rp-ml@rapid.lpt.fi
Sent: Wed, January 20, 2010 10:42:00 PM
Subject: [rp-ml] low-res scanning - a follow up

Dear All

Thanks to everyone for the very useful information in reply to my query. Rather than reply to everyone individually, perhaps it would be better to just clarify a little on the rp-ml: -

The responses have pretty much confirmed my original thinking and I now have some very good leads regarding possible technologies suitable for this application.

To those who offered scanning services, I have to decline. The reason is because the possible project would involve around 1000 stone items to be moved around the world. I'm sure you understand the problem.

The purpose would be to create a gallery where people can virtually manipulate the piece so a 3D model is a requirement rather than just a video. The other requirement is to have this in a very small file, less than 1Mb per object, which I still am not sure would be viable without significantly compromising the visual quality and accuracy.

If anyone has managed to get this far in my message, perhaps you can give me some further insight? What I want to know is what is the smallest 3D textured model you have managed to obtain? If you are able to send me an example (.obj or whatever) I would love to see it.

Regards

IG
Received on Fri Jan 22 20:11:02 2010