google/mannequinchallenge

Colmap Patch matching settings

mbaradad opened this issue · 2 comments

Hi,
I am trying to replicate your training method, but my results from colmap look more dense (though also noisier) than yours (example at the end). I assume this is caused by a different setting on the patch matching (I am just using the default one) and not only the MegaDepth post processing steps. Could you provide further details about the hyperparameters you used?
Thanks!

IggIqNXfu_U_119519400:
image
Yours:
image
Mine:
image

Hi, you could take a look at the documentation I wrote here for reproducing SfM+MVS results in our MC dataset: https://docs.google.com/document/d/1lWOcbLIeGGVVpjkGiMaq0zRVZvaBJnewRUPN2mdAD_A/edit?usp=sharing

Thanks, this is really useful!