fanwuyang/UMSN-Face-Deblurring
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
rajeevyasarla
2019-08-06 14:21:17 -04:00
committed by GitHub
parent 85d53eae36
commit 58e6924938
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+2 -2
View File
@@ -14,14 +14,14 @@ We propose a novel multi-stream architecture and training methodology that explo
4. CPU or NVIDIA GPU + CUDA CuDNN (CUDA 8.0)
## To test UMRL:
## To test UMSN:
1. Download test datasets provided the authors of Ziyi et al.
- https://sites.google.com/site/ziyishenmi/cvpr18_face_deblur
2. run test_data_generation.m
- It renames the files counting from 1, for example 000001.png
3. python test_face_deblur.py --dataroot ./facades/github/ --valDataroot <path_to_test_data> --netG ./pretrained_models/Deblur_epoch_Best.pth
## To train UMRL:
## To train UMSN:
1. Kernels are generated using,
- [Boracchi and Foi, 2012] Modeling the Performance of Image Restoration from Motion Blur Giacomo Boracchi and Alessandro Foi, Image Processing, IEEE Transactions on. vol.21, no.8, pp. 3502 - 3517, Aug. 2012,
- 25000 kernels with size ranging from 13 to 29 are generated and saved as ".mat" file