I've been struggling with larger objects (chairs) and smaller objects with lots of details (old 8mm cameras) so I tried something as simple as possible and started to find some success using a set of steampunk styled dice - here on Sketchfab
This upload is still pretty rough and only shared privately, but I wanted to share a post on finding a little bit of success and maybe inspire others to share tips on what they're discovering.
Some general feedback - I've had limited success with the auto-capture mode. Using the manual capture mode and applying basic overlapping principles works a lot better for me. I've found the auto-capture mode often goes haywire on me -- maybe it's the lighting, or maybe I'm just too unsteady, but I get a lot of blurry image warnings.
A question for the dev team - I've made over a dozen scans of various objects and I've got a stack of awful looking drafts inside Sketchfab. I presume the data is already available to the RealityScan dev team -- but I want to ask -- should beta testers share failed model attempts privately with the dev team to see our struggles? Or are you already seeing them?
I'm inclined to only show successful looking attempts in a forum like this, but I imagine the failures might be much more useful to the dev team.
Thanks for the amazing work. I hope this is helpful.
Please sign in to leave a comment.