Hi, I am a bit new to Reality Scan. I am working with a drone captured data set of a large vertical object. The AI masking feature creates a fantastic reconstruction of the tall vertical, but everything on the ground gets garbled. Is there a way to combine a masked point cloud and regular point cloud together? The regular reconstruction doesn’t come close to the quality of the vertical structure that is needed but it does have all the elements needed on the ground. Also is there a way to increase the tiling in reality scan? I see the model is in chunks, but wondered if there is a way to fine tune it into smaller sections. I apologize if these questions have been answered before, but couldn’t find exactly answers that fit my question when searching. Thanks!
The best answer I have found and have confirmed it worked on my test case is to create a regular project and align images. Then export out to .rsalign format. Create a new project and do the Ai masking, and align images. Then Drag the .rsalign export out from the folder into the ai project and align components. I ended up with a complete point cloud that took advantage of the masking and had the full capture. The end model was really clean. I have tried it larger projects, but with some successes and some failures. I hope this is a feature that will continue to evolve, because it can produce some amazing results on lattice work and complicated subjects normally difficult to get a reconstruction of.