search

kcheng1021 / GaussianPro

[ICML2024] Official code for GaussianPro: 3D Gaussian Splatting with Progressive Propagation
https://kcheng1021.github.io/gaussianpro.github.io/
MIT License
557 stars 38 forks source link

Error： ModuleNotFoundError: No module named 'pcpr' #4

Closed cv-lab-x closed 3 months ago

cv-lab-x commented 3 months ago

thanks for your great work, there's some errors when run the waymo demo, some files may missing when build Propagation ?

File "GaussianPro/utils/graphics_utils.py", line 16, in <module> import pcpr ModuleNotFoundError: No module named 'pcpr'

Looking forward to your reply，thanks! @kcheng1021

Disa-Kizonda commented 3 months ago

https://github.com/wuminye/PCPR worked for me

kcheng1021 commented 3 months ago

Apologies for this. The code about 'pcpr' was part of my experimental attempts. Please feel free to remove the relevant packages and codes directly.

cv-lab-x commented 3 months ago

Apologies for this. The code about 'pcpr' was part of my experimental attempts. Please feel free to remove the relevant packages and codes directly.

thanks for your quick reply, another Error: FileNotFoundError: [Errno 2] No such file or directory: './cache/propagated_depth/costs.dmb'

Looking forward for your reply, thanks! @kcheng1021

kcheng1021 commented 3 months ago

Does the path './cache/propagated_depth/' exist in your project? Could you offer the complete print about the error?

cv-lab-x commented 3 months ago

Does the path './cache/propagated_depth/' exist in your project? Could you offer the complete print about the error?

ye, the dir exists.

`Loading Test Cameras [02/03 19:59:21] Number of points at initialisation : 72378 [02/03 19:59:23] Training progress: 0%| | 0/30000 [00:00<?, ?it/s] [ITER 1] Evaluating test: L1 0.20549551770091057 PSNR 11.761059880256653 [02/03 19:59:27]

[ITER 1] Evaluating train: L1 0.232677760720253 PSNR 10.858474922180177 [02/03 19:59:31]

[ITER 1] Saving Gaussians [02/03 19:59:31] Training progress: 3%|█████▋ | 1010/30000 [01:21<34:26, 14.03it/s, Loss=0.3947600]terminate called after throwing an instance of 'cv::Exception' what(): OpenCV(4.5.4) ./modules/core/src/matrix_expressions.cpp:24: error: (-5:Bad argument) Matrix operand is an empty matrix. in function 'checkOperandsExist'

Aborted (core dumped) Traceback (most recent call last): File "train.py", line 372, in training(lp.extract(args), op.extract(args), pp.extract(args), args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from) File "train.py", line 129, in training propagated_depth, cost, normal = read_propagted_depth('./cache/propagated_depth') File "GaussianPro/utils/general_utils.py", line 244, in read_propagted_depth cost = readDepthDmb(os.path.join(path, 'costs.dmb')) File "GaussianPro/utils/general_utils.py", line 192, in readDepthDmb inimage = open(file_path, "rb") FileNotFoundError: [Errno 2] No such file or directory: './cache/propagated_depth/costs.dmb' Training progress: 3%|█████▋ | 1010/30000 [01:22<39:18, 12.29it/s, Loss=0.3947600] Looking for config file in output/gs_pro/cfg_args `

kcheng1021 commented 3 months ago

I haven't encountered this error on my server, and I guess it may be related to the specific versions of the environment. I will try to reproduce this error later. For now, I'll provide some information about the computer I'm testing on: Ubuntu 20.04.1 LTS, GeForce 3090. As for the versions of OpenCV, for opencv-python, our version is 4.9.0.80. For opencv for C++ in the Propagation module, our version is 4.2.0.

GongMingCarmen commented 3 months ago

@cv-lab-x "I encountered the same issue, how did you solve it?"

Valkyrie1215 commented 3 months ago

+1, there is no "costs.dmb"file in the './cache/propagated_depth',hasn't it been uploaded yet? image

cv-lab-x commented 3 months ago

@cv-lab-x "I encountered the same issue, how did you solve it?"

not yet

kcheng1021 commented 3 months ago

These files are generated during the execution of the propagation process rather than pre-existing. It's possible that an error occurred during the propagation steps. Could you please provide your hardware and environment information? I will try to reproduce the bug.

kcheng1021 commented 3 months ago

This issue about 'pcpr' has been solved. To help others better get the related information. I would like to merge your further questions with Issue #8.

cv-lab-x commented 3 months ago

This issue about 'pcpr' has been solved. To help others better get the related information. I would like to merge your further questions with Issue #8.

i have fixed the bugs, and add a PR to fixing it

kcheng1021 commented 3 months ago

This issue about 'pcpr' has been solved. To help others better get the related information. I would like to merge your further questions with Issue #8.

i have fixed the bugs, and add a PR to fixing it

Thanks for your debuging and help! But we found why the ref_id 0 is not be passed. Please refer to Issue https://github.com/kcheng1021/GaussianPro/issues/8