[question] How to use depth maps (png) from external source?

[ZeyadZewail]

i am trying to generate a 3D model from frames of a video alongside depth maps that have been generated for every frame using deep learning.

can anyone explain how i would go about doing this?

this is the project i am using for the depth maps https://roxanneluo.github.io/Consistent-Video-Depth-Estimation/

and here is the dataset if u wanna try https://www.dropbox.com/sh/zsvmbc5iy2br8ol/AAAcEo5M9KYBSN7aiAuSPttka?dl=0

i tried converting the png output depth maps into exr manually and renaming them but i am getting this error

[20:13:38.660663][info] Found 1 image dimension(s): [20:13:38.661655][info] - [1080x1920] [20:13:38.662656][info] Reading 'F:/GP/meshroom workspace/depth/46121852_depthMap.exr' x4downscale from file dimension - No 'AliceVision:downscale' metadata found. [20:13:38.671664][fatal] Scale of file: 'F:/GP/meshroom workspace/depth/46121852_depthMap.exr' is not uniform, check image dimension ratio.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[VihangMhatre]

Did you get an answer to your issue?

[natowi]

Here is a similar topic https://github.com/alicevision/Meshroom/issues/1361