evo Posted February 6, 2017 Report Share Posted February 6, 2017 Hi CHI, Could you all remind me again of the protocol for sphere selection? We're imaging a large paper document ~3x2' and the RTI turned out great, except I noticed in the RTIBulider that the sphere diameter was far below the 250pixels I'd written down in my notes when taking the course. The spheres selected for this project were very small given the thickness of the paper (so that only 1/3 would be in focus). Do we have an issue with insufficient resolution or sphere selection? Link to comment Share on other sites More sharing options...
cdschroer Posted February 6, 2017 Report Share Posted February 6, 2017 A couple of things to note regarding sphere size: 1. While the capture guide and our training talks about the pixels in the diameter of the sphere, RTIBuilder reports the radius - so you need to double it to get the diameter. 2. As you saw, the software will work if you are below the recommended number. The issue is that your light positions aren't as accurate, because you don't have a pixel for each degree on the sphere. For interpretive purposes in the viewer, you wouldn't be able to tell the difference. For some applications where normal accuracy is being quantified, it could make a difference. Carla Link to comment Share on other sites More sharing options...
evo Posted February 7, 2017 Author Report Share Posted February 7, 2017 Good to know, thanks for that. Could you recommend some strategies for increasing this? Should we use larger spheres and slightly elevate the paper document? Link to comment Share on other sites More sharing options...
cdschroer Posted February 7, 2017 Report Share Posted February 7, 2017 Hard to say without seeing your setup and the subject. Sometimes a larger sphere and just setting it below the edge will work fine. Sometimes you have to place it on the matte or similar area outside the area you are interested in at the same height. In that case you should set the spheres a bit away from the area you care about to reduce shadowing on the area. And finally, it is possible to get truncated spheres, you want about a 3rd of the sphere removed. We have some that we purchased ages ago. I don't have a link, and they were quite pricey, so we didn't feel we could put them in the kit. You might be able to find a machine shop that could modify some ball bearings for you, if you will be doing a lot of this kind of work. Good luck! Carla Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.