search

autonomousvision / sdfstudio

A Unified Framework for Surface Reconstruction
Apache License 2.0
1.98k stars 185 forks source link

Extracting MonoCues #19

Closed LownyCGI closed 1 year ago

LownyCGI commented 1 year ago

Having a trouble extracting MonoCues

It's just doesn't extract

Do i need to get camera poses beforehand?

image

niujinshuchong commented 1 year ago

Hi, I haven't tested it on the Windows system. But looks like the path is mixed with \\ and /. Could you make it consistent and try again?

LownyCGI commented 1 year ago

sorry. can this path be considered consistent? if so, same result

python scripts/datasets/extract_monocular_cues.py --task normal --img_path datatest/office06/PhotoSeq/ --output_path datatest/office06/resulted/ --omnidata_path ../.conda/envs/omnidata/Lib/site-packages/omnidata_tools/torch/ --pretrained_models ../.conda/envs/omnidata/Lib/site-packages/omnidata_tools/torch/pretrained_models/

niujinshuchong commented 1 year ago

Hi, in your screenshot, the path is sometimes C:\\ and in the end C:/. I am not sure whether this is the problem. I think you could try changing all of them to \\.

LownyCGI commented 1 year ago

i tried \\ it doubled up in quantity.

LownyCGI commented 1 year ago

image

Hi. I reinstalled sdfstudio and put omnidata-tools on top of it. and have been running into the same problem ever since

LownyCGI commented 1 year ago

Solved. incorrect filenames for images. Thanks to @joeylitalien

https://github.com/autonomousvision/sdfstudio/pull/22#issuecomment-1379291842