I’m working with a client that is using a skydio drone and an iphone 12 pro max to collect data for house models.
We’re having a terrible time matching these photos together. I simply cannot get the iphone photos to link, despite using great tie points. Looking at the tie points for the iphone component vs the drone component, they seem to be selected differently, as if there is some fundamental difference in how the pixels are generated.
He sent me the dataset today, and I’m going to fiddle with it a bunch, but I’m at a loss watching his workflow, and watching tie points fail.
Could you provide some screenshots of the resulting components? We have to check the capturing pattern of the drone and the phone.
There was likely not enough overlap between the drone and phone images. In that case, if it is possible, I would recommend adding more images.
If that is not possible, using control points can be your next option. But that can be time consuming, and it doesn’t guarantee a good result.
If you want, you can check out this video tutorial I recently recorded. I am processing an old chapel. I used a drone and a mirrorless camera. If you maintain high overlap between images, RC will merge everting automatically.
Hi Jakob
I’ll have to upload screenshots a bit later.
I’d estimate 80% or greater overlap on some of the iphone shots. I added control points, including RC-generated targets, on more than 20 images for each. It pulled in zero iPhone images, so I increased the control point weight all the way to 100,000. No luck. I tried with prior position toggled on and off. No luck. The upfront photos always make their own component, which I cannot merge with the drone component.
It feels broken. Nothing changes, regardless of control, settings, etc etc.