oxyhexagen

Hi,so how did you deal with it?

> Then, I did the evaluation of pose estimation on my dataset using your models. I found that the old model (outdoor_ds.ckpt) with temp_bug_fix=False works much better than the new...

Have you solved it yet? I googled everywhere and this is the only issue I found same with me

Thought i've just solved a similar problem [#1686](https://github.com/colmap/colmap/issues/1686#issuecomment-1311246225)

This isn't a critical issue requiring changes, just a minor note about a problem in this document. I happened to read from the consistency graph binary file saved by MVS...

> @vvvsice Good question. > > It's true that depths around the points are not same. The key idea is to compare patch similarity between different frames. You can also...

It's clear,thanks!