search

apchenstu / mvsnerf

[ICCV 2021] Our work presents a novel neural rendering approach that can efficiently reconstruct geometric and neural radiance fields for view synthesis.
MIT License
672 stars 84 forks source link

Instructions for custom data #27

Open reconlabs-sergio opened 2 years ago

reconlabs-sergio commented 2 years ago

Hi,

Thanks for this great code. I noted that there are no instructions for running this on custom data (the link is broken). Would you mind explaining what is the process to run this with custom data? Your example data has Camera, Depth and Rectified folders. Are all of these necessary? I couldn't find in your paper any indication on how these are generated.

Any help will be greatly appreciated!

apchenstu commented 2 years ago

The depth is not necessary for training, our depth is just for quantitative evaluation. Actually, you can convert your data into the DTU format if you are familiar with it. If not you can refer to the nerf++ project, I remember they have provided the calibration and conversion code. There are no instructions on custom data since I don't test it on custom data:)

G20202502 commented 2 years ago

Is there any portable way to remove depth map from the input? Now I want to generate my custom dataset, and I don't have the depth maps:( thanks a lot for any help~