Vox-Fusion

Vox-Fusion copied to clipboard

Hello author, this is a very great job. For beginners, this config file is missing causing a big barrier, hopefully the author can update it sometime!

LSK0821

https://github.com/zju3dv/Vox-Fusion/blob/282768d4c121ee41c6696dc123f3805482abdeeb/src/variations/render_helpers.py#L315 ``` return { "weights": weights, "color": rgb, "depth": depth, "z_vals": z_vals, "sdf": sdf, "weights": weights, "ray_mask": ray_mask, "raw": z_min if return_raw else None } ```

LZL-CS

I have different results when evaluating the trace, can you share the specific code of your trajectory evaluation. Thank you

Aurora-chevalier

traj.txt

3

hello,I have a question about traj.txt,the traj.txt is gt or other?

xiaoshumiao123

I record a data sequence by myself and run the code. After processing about 800 frames, the following error appears. It seems that the dim of `map_states["voxel_vertex_idx"]` and `map_states["voxel_center_xyz"]` exceeds...

jarvishou829

CUDA out of memory.

5

I am trying to run scannet/scene0059, but got cuda out of memory error. Here is the error message: ``` home/slam/.conda/envs/ngp_pl/lib/python3.8/multiprocessing/process.py", line 315, in _bootstrap self.run() File "/home/slam/.conda/envs/ngp_pl/lib/python3.8/multiprocessing/process.py", line 108, in...

Xiaxia1997

The reason is that torch will use `ninja` to build if it's installed and caused a header path issue. To solve the issue, replacing the line in `third_party/sparse_octree/setup.py` and `third_party/sparse_voxels/setup.py`...

TomCC7

Producer process has been terminated before all shared CUDA tensors released

8

#7 #6

garylidd

Update key-frame selection strategy

1

#8

garylidd

#4

garylidd