1: in your scripts/shells/train.sh，--init_points_filename points3D_withlidar.txt. But in 8398516118967750070_3958_000_3978_000/colmap/sparse/0, there is only points3D.bin, points3D_clean.bin, points3D1.bin, points3D3.bin
2: when I use your preprocess dataset: 8398516118967750070_3958_000_3978_000, execute ./scripts/shells/train.sh,
`data_root=$1
cuda_id=$2
Step (% Done) Train Iter (time) ETA (time) Train Rays / Sec
-----------------------------------------------------------------------------------
680 (2.27%) 154.155 ms 1 h, 15 m, 19 s 4.16 M
690 (2.30%) 156.785 ms 1 h, 16 m, 35 s 4.10 M
700 (2.33%) 160.737 ms 1 h, 18 m, 29 s 4.02 M
----------------------------------------------------------------------------------------------------
Viewer running locally at: http://localhost:7007 (listening on 0.0.0.0)
Printing profiling stats, from longest to shortest duration in seconds
Trainer.train_iteration: 0.1052
VanillaPipeline.get_train_loss_dict: 0.1000
Traceback (most recent call last):
File "envs/street-gaussians-ns/bin/sgn-train", line 8, in <module>
sys.exit(entrypoint())
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/scripts/train.py", line 262, in entrypoint
main(
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/scripts/train.py", line 247, in main
launch(
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/scripts/train.py", line 189, in launch
main_func(local_rank=0, world_size=world_size, config=config)
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/scripts/train.py", line 100, in train_loop
trainer.train()
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/engine/trainer.py", line 261, in train
loss, loss_dict, metrics_dict = self.train_iteration(step)
File "/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/utils/profiler.py", line 112, in inner
out = func(*args, **kwargs)
File "street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/engine/trainer.py", line 496, in train_iteration
_, loss_dict, metrics_dict = self.pipeline.get_train_loss_dict(step=step)
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/utils/profiler.py", line 112, in inner
out = func(*args, **kwargs)
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/pipelines/base_pipeline.py", line 301, in get_train_loss_dict
model_outputs = self._model(ray_bundle) # train distributed data parallel model if world_size > 1
File "street-gaussians-ns/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "envs/street-gaussians-ns/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
return forward_call(*args, **kwargs)
File "envs/street-gaussians-ns/lib/python3.8/site-packages/nerfstudio/models/base_model.py", line 143, in forward
return self.get_outputs(ray_bundle)
File "street-gaussians-ns/street_gaussians_ns/sgn_splatfacto_scene_graph.py", line 363, in get_outputs
out = super().get_outputs(camera)
File "/street-gaussians-ns/street_gaussians_ns/sgn_splatfacto.py", line 861, in get_outputs
self.xys, self.depths, self.radii, self.conics, _, self.num_tiles_hit, _ = project_gaussians( # type: ignore
File "envs/street-gaussians-ns/lib/python3.8/site-packages/gsplat/project_gaussians.py", line 60, in project_gaussians
assert (quats.norm(dim=-1) - 1 < 1e-6).all(), "quats must be normalized"
AssertionError: quats must be normalized
1: in your scripts/shells/train.sh，--init_points_filename points3D_withlidar.txt. But in 8398516118967750070_3958_000_3978_000/colmap/sparse/0, there is only points3D.bin, points3D_clean.bin, points3D1.bin, points3D3.bin
2: when I use your preprocess dataset: 8398516118967750070_3958_000_3978_000, execute ./scripts/shells/train.sh,
`data_root=$1 cuda_id=$2
mkdir -p output/
CUDA_VISIBLE_DEVICES=$cuda_id sgn-train street-gaussians-ns \ --experiment_name street-gaussians-ns \ --output_dir output/ \ --vis viewer \ --viewer.quit_on_train_completion True \ colmap-data-parser-config \ --data $data_root \ --colmap_path colmap/sparse/0 \ --load_3D_points True \ --max_2D_matches_per_3D_point 0 \ --undistort True \ --segments-path segs \ --filter_camera_id 1 \ --init_points_filename points3D.bin`
error is:
when I print quats，there are some "nan" in quats.