search

cuiziteng / ICCV_MAET

🌕 [ICCV 2021] Multitask AET with Orthogonal Tangent Regularity for Dark Object Detection. A self-supervised learning way for low-light image object detection.
Apache License 2.0
142 stars 14 forks source link

Pre-trained models #2

Closed RanBenIzhak closed 2 years ago

RanBenIzhak commented 2 years ago

Excellent work! Thanks you for providing the code.

In the pre trained models section, the google drive links for: MAET-COCO and MAET-EXDark (77.7)

Point to the same file.

could you please fix the links for the weights on google drive?

cuiziteng commented 2 years ago

Hi, thanks for interest in our work !

I've fixed that error link on google drive, any question please drop here~

RanBenIzhak commented 2 years ago

Thank you for the fast reply!

I did not manage to load the pre-trained model, with the following error code -

(tools\test.py - line 164 - checkpoint = load_checkpoint(model, args.checkpoint, map_location='cpu')

Traceback (most recent call last): File "/home/.../.conda/envs/MAET/lib/python3.7/site-packages/torch/serialization.py", line 585, in load return _legacy_load(opened_file, map_location, pickle_module, pickle_load_args) File "/home/.../.conda/envs/MAET/lib/python3.7/site-packages/torch/serialization.py", line 755, in _legacy_load magic_number = pickle_module.load(f, pickle_load_args) _pickle.UnpicklingError: A load persistent id instruction was encountered, but no persistent_load function was specified.

For the model argument i was giving the "MAET-EXDark/data.pkl" path. Any idea on how to solve the issue?

Thanks!

cuiziteng commented 2 years ago

Hi, Lzhak~ here are two things you need to do after clone this project :

cd "your_project_path" --- and then

  1. pip install -r requirements/build.txt
  2. pip install -v -e .

see (here)[https://mmdetection.readthedocs.io/en/latest/get_started.html#installation] for more details.

After trying these two steps, I think it may work~

cuiziteng commented 2 years ago

I've upload this in readme.md.