Open Wangfeng66999 opened 1 week ago
Hi, thanks for your interest!
Please check if the mesh
here returns a valid result. I also recommend verifying the related variables, such as sample
and self.data
. If everything is correct, you should get results similar to the image below:
Additionally, please ensure that the json_file
here is functioning correctly.
Hi, I followed your debugging process, and when I print(sample):
Hi @xXuHaiyang, thanks for open sourcing your great work! I ran into a problem while running "preprocess_pix3d.py", can you help me?
Found 3839 samples for category chair Using 3839 samples for all 0%| | 1/3839 [00:00<03:18, 19.30it/s] Traceback (most recent call last): File "/home/robot/Project/BDM/experiments/data/Pix3D/preprocess_pix3d.py", line 98, in
dataset[i]
File "/home/robot/Project/BDM/experiments/data/Pix3D/preprocess_pix3d.py", line 63, in getitem
pointcloud = sample_points_from_meshes(mesh, self.sample_size).squeeze()
File "/home/robot/miniconda3/envs/point/lib/python3.9/site-packages/pytorch3d/ops/sample_points_from_meshes.py", line 65, in sample_points_from_meshes
raise ValueError("Meshes are empty.")
ValueError: Meshes are empty.