Sorry I had to break down and create a new post on this subject...I've read all the other posts and none of them provide answers to what seems like should be a very straightforward workflow. I am scanning trees and their surrounding environment. When I take the photos, I place 3 markers level with one another into the environment and take measurements between them. Based on previous workflows with different software, I would import a photogrammetry scan into 3ds max and do the scaling and aligning there based on my measurements of those markers, but Reality Capture seems like it has the tools to do this natively, and I'd much prefer to do it here.
Here's my process, followed by what isn't working as expected:
- Add imagery Inputs
- Process Align Images
- Calculate model
- Select & Filter out large triangles (would be great if I could just turn off an option for the software to create a "water tight" mesh...I don't need this functionality and makes me do extra steps)
- Process Texture
- Create 3 control points in the 3D view (clicking directly on my physical markers)
- Define Distance between 2 of those control points*
*this is where the process breaks down. Initially it wouldn't give me a way to define the distance myself, telling me I needed to edit the suggested points in the 2D photos. (I'm not sure why this is necessary since I can see that the control points are in the right location in the 3D view.) I went ahead and manually moved or created the control points in a bunch of the 2D photos and tried to define the distance again. Defining a distance shows a "calculated distance", which is obviously not correct. When I type in my known measurement into "Defined distance" and hit Update, I get the following error:
"Updating Constraints (Draft) - Operation Failed. No component with valid ground control constraints found."
I'm sure I'm just missing a step that isn't clear. It sure would be nice if this process was more intuitive. The workflow process takes me through each step in an intelligent way, but once I have my model, it would make sense to simply click on my 3 markers to make control points, click on all 3 points to establish my know level, then click from 1 point to another and enter my known distance. Feels like 8 clicks and I'd have a perfectly scaled and oriented scene. But all this finagling is very challenging. I'll stop the rant for now :) and just hope someone can chime in with the step of the process I'm missing to be able to update an incorrect "calculated distance" to a correct "defined distance". Thanks so much!
Here's a screenshot of my scene:
Please sign in to leave a comment.