search

princeton-vl / infinigen

Infinite Photorealistic Worlds using Procedural Generation
https://infinigen.org
BSD 3-Clause "New" or "Revised" License
5.32k stars 455 forks source link

Blender ground truth not generated for video #152

Open aenoca opened 11 months ago

aenoca commented 11 months ago

Describe the bug

A clear and concise description of what the bug is.

The ground truth of depth and semantic segmentation is not generated using the following command, though it is specified. Furthermore, only frames 17 to 24 are rendered and the others are not.

Steps to Reproduce

Simply run the following command.

What version of the code were you using?

''' What is the output of git log -n 1? '''

What command did you run?

python -m tools.manage_datagen_jobs --output_folder outputs/my_video5 --num_scenes 1 --pipeline_configs local_64GB.gin monocular_video.gin blender_gt.gin cuda_terrain.gin  --configs desert.gin simple.gin --pipeline_overrides iterate_scene_tasks.frame_range=[1,25] -p camera.camera_pose_proposal.altitude=2

What are your FULL output logs?

If this is your first time running Infinigen, what are the full output logs of install.sh ?**

Platform

Additional context

Add any other context about the problem here.

araistrick commented 11 months ago

Could you post all the relevant logs and the version of the code you are using?

We also just released some pre-generated videos if that would interest you https://github.com/princeton-vl/infinigen/blob/main/docs/PreGeneratedData.md.