sfm issues

Question

sfm issues

Closed this issue 2 years ago · 9 comments

Hello, thank you for sharing your code, I want to ask you a question, when I run the run.py file, I run to (match_worker_ray_wrapper pid=31722) Global seed set to 666 here, it keeps stuck, by the way my DeepLM is not installed successfully, please ask this question How to solve this problem?

(oneposeplus) lgq@lgq-B560M-AORUS-PRO-AX:/mnt/500_disk/OnePose_Plus_Plus$ python run.py +preprocess=sfm_inference_lowtexture.yaml use_local_ray=True
run.py:398: UserWarning:
The version_base parameter is not specified.
Please specify a compatability version level, or None.
Will assume defaults for version 1.1
@hydra.main(config_path="configs/", config_name="config.yaml")
/home/lgq/anaconda3/envs/oneposeplus/lib/python3.8/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config.yaml': Defaults list is missing _self_. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/lgq/anaconda3/envs/oneposeplus/lib/python3.8/site-packages/hydra/_internal/hydra.py:119: UserWarning: Future Hydra versions will no longer change working directory at job runtime by default.
See https://hydra.cc/docs/1.2/upgrades/1.1_to_1.2/changes_to_job_working_dir/ for more information.
ret = run_job(
2023-03-25 11:30:45.174 | INFO | main:sfm:32 - Process all objects in directory:/mnt/500_disk/OnePose_Plus_Plus/data/datasets/lowtexture_test_data/lowtexture_test_data, process: 1 sequences
2023-03-25 11:30:45.176 | INFO | main:sfm_worker:92 - Worker: 0 will process: ['0700', '0701', '0702', '0703', '0706', '0707', '0708', '0710', '0712', '0713', '0714', '0715', '0718', '0719', '0720', '0721', '0722', '0723', '0724', '0725', '0726', '0727', '0728', '0729', '0730', '0731', '0732', '0733', '0734', '0735', '0736', '0737', '0740', '0741', '0742', '0743', '0744', '0745', '0746', '0748'], total: 40 objects
0%| | 0/40 [00:00<?, ?it/s]2023-03-25 11:30:45.177 | INFO | main:sfm_worker:97 - Processing /mnt/500_disk/OnePose_Plus_Plus/data/datasets/lowtexture_test_data/lowtexture_test_data/0700-toyrobot-others toyrobot-1.
2023-03-25 11:30:47.499 | INFO | main:sfm_core:193 - Keypoint-Free SfM coarse reconstruction begin...
(match_worker_ray_wrapper pid=31640) Global seed set to 6660/965 [00:00<?, ?it/s]
(match_worker_ray_wrapper pid=31642) Global seed set to 666
(match_worker_ray_wrapper pid=31641) Global seed set to 666
(match_worker_ray_wrapper pid=31722) Global seed set to 666

Answer 1 · 2023-03-25T03:59:00.000Z

Hi, please refer to this issue to see whether my reply can solve your problem!

Answer 2 · 2023-03-25T04:07:13.000Z

Thank you, thank you so much!

Answer 3 · 2023-03-25T04:38:22.000Z

@hxy-123 Hi, I'd like to ask you the following, what is the difference in the result between use_local_ray=True and use_local_ray=False in the command, because now I'm stuck at 93% when matching image pairs with use_local_ray=True, but it works fine with use_local_ray=False.

Answer 4 · 2023-03-25T04:44:37.000Z

@hxy-123 Hi, I'd like to ask you the following, what is the difference in the result between use_local_ray=True and use_local_ray=False in the command, because now I'm stuck at 93% when matching image pairs with use_local_ray=True, but it works fine with use_local_ray=False.

We use ray for parallelized matching during SfM which will be faster. The results will be the same. Getting stuck during parallelized matching may be due to the CUDA OOM since we use four workers in a single GPU.

Answer 5 · 2023-03-25T04:46:33.000Z

Thank you for your answer, you are the most prompt answer I have encountered, I wish you more high level papers

Answer 6 · 2023-03-25T04:49:50.000Z

Hi, I'm sorry to ask, but did you encounter the problem of no model name BACore when installing DeepLM? How to solve this problem @hxy-123

Answer 7 · 2023-03-25T04:55:43.000Z

Hi, I'm sorry to ask, but did you encounter the problem of no model name BACore when installing DeepLM? How to solve this problem @hxy-123

Is DeepLM properly installed or can it be successfully imported? Without DeepLM you can still run our code with our first-order solver.

Answer 8 · 2023-03-25T05:03:17.000Z

Thanks for your reply

Answer 9 · 2024-09-03T03:43:40.000Z

Hello, thank you for sharing your code, I want to ask you a question, when I run the run.py file, I run to (match_worker_ray_wrapper pid=31722) Global seed set to 666 here, it keeps stuck, by the way my DeepLM is not installed successfully, please ask this question How to solve this problem?

(oneposeplus) lgq@lgq-B560M-AORUS-PRO-AX:/mnt/500_disk/OnePose_Plus_Plus$ python run.py +preprocess=sfm_inference_lowtexture.yaml use_local_ray=True
run.py:398: UserWarning:
The version_base parameter is not specified.
Please specify a compatability version level, or None.
Will assume defaults for version 1.1
@hydra.main(config_path="configs/", config_name="config.yaml")
/home/lgq/anaconda3/envs/oneposeplus/lib/python3.8/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config.yaml': Defaults list is missing _self_. See https://hydra.cc/docs/1.2/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/lgq/anaconda3/envs/oneposeplus/lib/python3.8/site-packages/hydra/_internal/hydra.py:119: UserWarning: Future Hydra versions will no longer change working directory at job runtime by default.
See https://hydra.cc/docs/1.2/upgrades/1.1_to_1.2/changes_to_job_working_dir/ for more information.
ret = run_job(
2023-03-25 11:30:45.174 | INFO | main:sfm:32 - Process all objects in directory:/mnt/500_disk/OnePose_Plus_Plus/data/datasets/lowtexture_test_data/lowtexture_test_data, process: 1 sequences
2023-03-25 11:30:45.176 | INFO | main:sfm_worker:92 - Worker: 0 will process: ['0700', '0701', '0702', '0703', '0706', '0707', '0708', '0710', '0712', '0713', '0714', '0715', '0718', '0719', '0720', '0721', '0722', '0723', '0724', '0725', '0726', '0727', '0728', '0729', '0730', '0731', '0732', '0733', '0734', '0735', '0736', '0737', '0740', '0741', '0742', '0743', '0744', '0745', '0746', '0748'], total: 40 objects
0%| | 0/40 [00:00<?, ?it/s]2023-03-25 11:30:45.177 | INFO | main:sfm_worker:97 - Processing /mnt/500_disk/OnePose_Plus_Plus/data/datasets/lowtexture_test_data/lowtexture_test_data/0700-toyrobot-others toyrobot-1.
2023-03-25 11:30:47.499 | INFO | main:sfm_core:193 - Keypoint-Free SfM coarse reconstruction begin...
(match_worker_ray_wrapper pid=31640) Global seed set to 6660/965 [00:00<?, ?it/s]
(match_worker_ray_wrapper pid=31642) Global seed set to 666
(match_worker_ray_wrapper pid=31641) Global seed set to 666
(match_worker_ray_wrapper pid=31722) Global seed set to 666

here：
#5 (comment)
my code has worked