SegAnyGAussians icon indicating copy to clipboard operation
SegAnyGAussians copied to clipboard
verified publisher icon Jumpat

About the preprocessing of the replica dataset

Open MajorTom0319 opened this issue 8 months ago • 2 comments

HI, I noticed that you are using this dataset, and i was wondering how could I get a colmap image sequence for each scene?

MajorTom0319 avatar Apr 18 '25 02:04 MajorTom0319

Hi, I'm not sure about what dataset do you mean. If you are asking about the lerf dataset, they used the nerf studio to convert the camera poses collected by specific APP to colmap form. You need to refer to NerfStudio for more details.

Jumpat avatar Apr 27 '25 08:04 Jumpat

Hi, I'm not sure about what dataset do you mean. If you are asking about the lerf dataset, they used the nerf studio to convert the camera poses collected by specific APP to colmap form. You need to refer to NerfStudio for more details.

I am referring to the Facebook replica-dataset. The official dataset does not seem to support 3DGS. I found the rendered images obtained in semanticNERF. Can I use the rendered images for experiments? Or did you do any preprocessing on this dataset?

THANKS!!!

MajorTom0319 avatar May 06 '25 08:05 MajorTom0319