[CVPR 2024] Official PyTorch implementation of SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering
I created a similar .json with the content {"./data/bicycle/":"./output/"}. But I encountered the following error when running it.
Binding radiance cloud to surface mesh...
Traceback (most recent call last):
File "/media/jinshan/data/cys/SuGaR/metrics.py", line 344, in
refined_sugar = SuGaR(
File "/media/jinshan/data/cys/SuGaR/sugar_scene/sugar_model.py", line 329, in init
scales = scales.clamp_min(0.0000001).reshape(len(faces_verts), -1, 1).expand(-1, self.n_gaussians_per_surface_triangle, 2).clone().reshape(-1, 2)
RuntimeError: cannot reshape tensor of 0 elements into shape [0, -1, 1] because the unspecified dimension size -1 can be any value and is ambiguous
My running command is python metrics.py --scene_config ./output/camerasss.json -r "density" , do you know what the reason is? How is your .json file setting? Thank you.
I created a similar .json with the content {"./data/bicycle/":"./output/"}. But I encountered the following error when running it. Binding radiance cloud to surface mesh... Traceback (most recent call last): File "/media/jinshan/data/cys/SuGaR/metrics.py", line 344, in refined_sugar = SuGaR( File "/media/jinshan/data/cys/SuGaR/sugar_scene/sugar_model.py", line 329, in init scales = scales.clamp_min(0.0000001).reshape(len(faces_verts), -1, 1).expand(-1, self.n_gaussians_per_surface_triangle, 2).clone().reshape(-1, 2) RuntimeError: cannot reshape tensor of 0 elements into shape [0, -1, 1] because the unspecified dimension size -1 can be any value and is ambiguous My running command is python metrics.py --scene_config ./output/camerasss.json -r "density" , do you know what the reason is? How is your .json file setting? Thank you.