Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

question about setting blender #4

Closed
amiltonwong opened this issue Sep 30, 2022 · 11 comments
Closed

question about setting blender #4

amiltonwong opened this issue Sep 30, 2022 · 11 comments

Comments

@amiltonwong
Copy link

amiltonwong commented Sep 30, 2022

Hi, @SteveJunGao ,

According to the steps, the following two lines are just used for ensuring the existence of pip and numpy:

./python3.7m -m ensurepip
./python3.7m -m pip install numpy 

Do we need to specify the location of bpy module, as it is needed here, when executing render_all.py ?

When I execute

python render_all.py --save_folder /media/root/data2/ShapeNet/render_view --dataset_folder /media/root/data2/ShapeNet/ShapeNetCore.v1 --blender_root /root/Downloads/blender-2.90.0-linux64/2.90/python/bin

I got the following error:

sh: 1: /root/Downloads/blender-2.90.0-linux64/2.90/python/bin: Permission denied

Any hints to solve this issue?

Thanks~

@SteveJunGao
Copy link
Collaborator

Hi @amiltonwong ,

Regarding the bpy, you don't need to specify the location for it, as the script render_shapenet.py is running inside Blender's python, and it will find the bpy automatically.

Regarding the error :

sh: 1: /root/Downloads/blender-2.90.0-linux64/2.90/python/bin: Permission denied

It seems you don't have access to the blender file you downloaded, can you double check whether you have access to it?

Also, to see more error message, you can remove >> tmp.out from the command in this line and it will print all the output from blender.

Hope this can help!

@amiltonwong
Copy link
Author

Hi, @SteveJunGao ,

Thanks for your reply. I've fixed the problem and figured out it's the wrong configure in my blender.

One more question, does the scales here have some typos? Two 0.9 scale values occur.

Thanks~

@saisurbehera
Copy link

saisurbehera commented Oct 1, 2022

How did you get it working @amiltonwong ? How did you fix the wrong configure ?

@SteveJunGao
Copy link
Collaborator

@amiltonwong the scale is correct, we normalize all the chairs to the length of 0.7, car & motorbikes to the scale of 0.9, to get the best view of the object when rendering it.

@SteveJunGao
Copy link
Collaborator

I close the issue now, if have any updates or further questions related to this issue, feel free to re-open it!

Thanks for trying the code!

@saisurbehera
Copy link

Do you know how much time this process takes? I am seeing about 14612 blender runs.

@SteveJunGao
Copy link
Collaborator

It depends on the capacity of your GPU card ;)

You can compute the time by (TIME_PER_OBJECT * N_OBJECT)

@saisurbehera
Copy link

image

It is only using a small parts of the GPU

@SteveJunGao
Copy link
Collaborator

Yes, the rendering is not GPU intensive. You can also run multiple processes in one GPU such that your GPU utilization can be higher

Note that, you only need to render the dataset once, after that you can train many different models using this rendered dataset.

@Freja71122
Copy link

--blender_root /root/Downloads/blender-2.90.0-linux64/2.90/python/bin

--blender_root should not be xxx/blender-2.90.0-linux64/2.90/python/bin, it should be xxx/blender-2.90.0-linux64/blender

@jpainam
Copy link

jpainam commented Jan 21, 2024

I got 14611 items preprocessed. and not 14612, one item is missing and I've opened an issue about it at #154

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants