Skip to content
/ GPEMSR Public

Official PyTorch code for Learning Large-Factor EM Image Super-Resolution with Generative Priors (GPEMSR, CVPR2024)

Notifications You must be signed in to change notification settings

jtshou/GPEMSR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 

Repository files navigation

Learning Large-Factor EM Image Super-Resolution with Generative Priors (GPEMSR)

Accepted by CVPR-2024
Jiateng Shou, Zeyu Xiao, Shiyu Deng, Wei Huang, Peiyao Shi, Ruobing Zhang, Zhiwei Xiong, Feng Wu*
MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition, University of Science and Technology of China
Institute of Artificial Intelligence, Hefei Comprehensive National Science Center
Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences
*Corresponding Author

Code instruction

We provide code for two dataset separately.
Code for AC4 dataset is coming soon.

Training

Three stages of our model are trained separately.
We provide all the training code and configuration files.

Testing

CREMI

Please put all the trained models in the 'pre-train_model' folder.

Output test images

python output_GPEMSR.py -opt option/output_GPEMSR_x8.yml # for 8x EMSR  
python output_GPEMSR.py -opt option/output_GPEMSR_x16.yml # for 16x EMSR

Visual metrics

Coming soon.

Segmentation metrics

Please put the pre-trained segmentation models in the 'pre-train_model' folder.

Superhuman MALA
GoogleDrive GoogleDrive
cd ./inference_code
python inference_seg.py -c config/seg_x8_superhuman.yaml # for 8x EMSR segmentation using superhuman model
python inference_seg.py -c config/seg_x8_MALA.yaml # for 8x EMSR segmentation using MALA model
python inference_seg.py -c config/seg_x16_superhuman.yaml # for 16x EMSR segmentation using superhuman model
python inference_seg.py -c config/seg_x16_MALA.yaml # for 16x EMSR segmentation using MALA model

Pre-trained models

We provide pre-trained models for all three stages.
We also provide pre-trained VGG model and SPyNet model.

CREMI AC4
GoogleDrive Coming Soon

16x EMSR

8x EMSR

Contact

If you have any problem with the released code, please contact me by email (shoujt@mail.ustc.edu.cn).

About

Official PyTorch code for Learning Large-Factor EM Image Super-Resolution with Generative Priors (GPEMSR, CVPR2024)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published