search

facebookresearch / DeepSDF

Learning Continuous Signed Distance Functions for Shape Representation
MIT License
1.4k stars 258 forks source link

Issue with pre-processing #44

Open Xingang1990 opened 4 years ago

Xingang1990 commented 4 years ago

When I ran the following command, I got warnings that no mesh found for instance.

Command python preprocess_data.py --data_dir data --source [...]/ShapeNetCore.v2/ --name ShapeNetV2 --split examples/splits/sv2_sofas_train.json --skip

Warnings DeepSdf - WARNING - No mesh found for instance 1037fd31d12178d396f164a988ef37cc DeepSdf - WARNING - No mesh found for instance 104256e5bb73b0b719fb4103277a6b93 DeepSdf - WARNING - No mesh found for instance 1053897adff12c7839c40eb1ac71e4c1 ... ... ... ... DeepSdf - WARNING - No mesh found for instance cc5f1f064a1ba342cbdb36da0ec8fda6

I downloaded the shapeNetCore.v2. And there are meshes for those instances. But they are not found. Does anyone encounter the same issue? Could you let me know how to fix this problem?

balamurugan15 commented 2 years ago

I'm facing the same issue. Any solution?

zhangshengjun2019 commented 2 years ago

I find a way that might solve this problem. You can add print(shape_dir) in find_mesh_in_directory data.py. And you will get the dir of the data. Then you can know whether such data_dir exists in your data_source or not. I didn't have mine in the data_source, so I got the same error. After I add the data, the preprocess_data.py can run. I hope this will help you.