CloudCompare forum

Hi there,
I have a feeling this is a stupidly simple problem, but I can't figure it out.
I have 7 point clouds of the same river bank and have opened them (.pts files) in CC. They all align really well so I'm very happy with them.However, I did some roughness analysis on one of the clouds, testing a number of different kernel sizes, and then saved the cloud as a .bin file, but when I open that .bin file it no longer aligns with any of the original .pts files. I've tried re-aligning the scans using the manual registration tools but I'm concerned that will affect the error values for any subsequent analysis (M3C2 is next on my list) and because they are very complex and rapidly changing environments I'm not sure I can even manually register them anywhere near as accurately as they were previously.
I have tried saving other clouds as .bin files and seeing if they 'shift' similarly, so that maybe the .bin files will align with each other but not with the original .pts files, however the second scan that I saved as a .bin file does still align with the original .pts files.

Has anyone experienced anything similar and maybe have an idea of how I can fix this?

Thanks
Becky

Hi Becky,
after aligning the pointcloud, the one saved pointcloud (.bin) has the transformation values applied while all the original pts files do not. You will have to realign all the clouds again, since cloud-compare does not just automatically overwrite the original files after alignment.

It would be best if you save all pointclouds in a single .bin file; this will ensure that the transformation (alignment) will be kept.
Cheers.

Yes. Pay attention also to an eventual 'Global shift' that may be saved in the BIN file. In this case, it will be automatically re-used when loading this file. But for the other files, CC may suggest another shift so you have to make sure it's the same (by manually entering the X and Y values ;).

Anyway, as Aarie noted, it's a better idea to save everything as a BIN file (ideally just one) instead of mixing the formats.

Ahh, I understand now. I had not been paying attention to the shift values.
Thank you for your replies.
Best wishes