I`m thinking that you should run the pictures into PMSV2. The GCP markers will show up there.

Hi Mark, I would very much like to compare notes. I think we've been exchanging ideas over at the KAP website. I am very impressed with the work you have been doing. I hope we can keep the conversation going. Regards-Nathan

Nathan, we should compare notes sometime. Your work is similar to some I've done in the highlands of Ecuador. -Mark

Hello Nathan, this is jimcseke. I am not sure if you remember me, but a while back you and I had a discussion in the Photosynth lounge about edit images for an improved point cloud. I have been away from the actual synthing for about 2 months now, and since you and I seemed to share an interest in accuracy, density, and editability of the point cloud, I though you might find the discussion I just started interesting and useful. try to read it in it's entirety as most of what is written is relevant. http://photosynth.net/discussion.aspx?cat=6b63cb81-8b57-4d5d-a978-41d5509bf59a&dis=c13fd173-a14e-4edf-8350-a3c28d6fb859 Have a great weekend, Sincerely Jim

You have got to check out this synth http://photosynth.net/view.aspx?cid=061d8fe5-dfc9-493a-9a45-80aae4d44f76 and my comments in the Photosynth Lounge. In regards to improving the pointcloud. And check out some of my synth which relate to the pointcloud improvement issue. I think (I know) I may have found a way to really improve pointclouds(images and accuracy)without taking more pics. Trust me on this one. I call it the "Jim's Zoom Cheat" or the "Cseke Cheat Zoom" (I need to leave some kind of legacy : )

Hi Jim, Sir_Ivar does wonderful work. I have been focusing on getting the point clouds into real world coordinates. I am extremely encouraged to hear Sir_Ivar report that synth is accurate. I am trying to figure out how to bring the clouds into a coordinate system that I can compare with other more traditional mapping methods (GPS & total station) so that I can have an empirical accuracy assessment. In the mean time, I'm glad you enjoyed this synth. It was made with a pole mount and a Canon G11. These things are amazingly quick to produce. If they are accurate, which I suspect they are, then this will be a major boost to archaeological field work. Thanks for your positive words. Warm regards, Nathan

Wow ! This is one of the great uses of Photosynth. I am assuming you have viewed some of sir_ivar's synths. One of my favorites of his is the Star Cave. I commented to him on how archaeology and Photosynth work beautifully together. In one of his comments he mentioned how accurate Photosynth was. Accurate information in position and size etc... of your subjects is extremely important. The combination of synthed photo, pointcloud, and viewer interaction create an informative and beautiful subject representation for people in your field and also laypersons like me. I hope you continue using photosynth and displaying your finished works. There are a lot of non-archeologial type people out there who love this type of subject. If you have the time, how is your post-processing experiment coming along? What if anything, have you found to improve any aspect of your synths?

Dear Madeeds, Thank you very much for these wonderful ideas, this is just the kind of feedback I am hoping for. I will try out your suggestions post the results. I think this technology holds great promise for archaeological documentation. Regards, Nathan

Very cool idea. Consider using a design that's more intricate than four squares of a checker board. With this design, you only get a few features at the corners. If it were more complicated, you could get more features. If you just did nine red spots on a yellow background, it might work better. You might also consider making each of the four targets use a different accent color so that in your post-processing you can identify them algorithmically.

Thank you very much for your suggestions regarding the cones. For the tests I did today, I drew on some lines and small hash marks. Scuffing the cones occurred to me. I'll give that a try. Thank you for the feedback. This is very helpful.

If you were hoping for pointclouds of the cones, it would have been better if they were dusty or scuffed up in addition to you getting the camera closer to their surfaces. You've made a perfect example of what I was going to write in response to your question on the forum, though: circling a spot with a camera can't hardly be beat for generating dense pointclouds.