From ddc8f7dd8a0d3f5ab66816f35dfb51030393e23e Mon Sep 17 00:00:00 2001 From: rajeevyasarla Date: Sat, 3 Aug 2019 11:23:08 -0400 Subject: [PATCH] UMSN Files --- README.md | 6 +++++- train_face_deblur.py | 2 +- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 5bd1581..a759036 100644 --- a/README.md +++ b/README.md @@ -21,7 +21,11 @@ We propose a novel multi-stream architecture and training methodology that explo 3. python test_face_deblur.py --dataroot ./facades/github/ --valDataroot --netG ./pretrained_models/Deblur_epoch_Best.pth ## To train UMRL: -python train_face_deblur.py --dataroot --valDataroot ./facades/github/ --exp ./face_deblur --batchSize 10 +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 +2. Clean face images from Helen and CelebA are aligned and used as input to train UMSN +3. python train_face_deblur.py --dataroot --valDataroot ./facades/github/ --exp ./face_deblur --batchSize 10 - input should be clean image. blurry images for training are generated by the code it self. diff --git a/train_face_deblur.py b/train_face_deblur.py index e781dc6..cab574c 100644 --- a/train_face_deblur.py +++ b/train_face_deblur.py @@ -356,7 +356,7 @@ for epoch in range(opt.niter): #x1 = xmask1*class_msk1*x_hat+(1-xmask1)*class_msk1*target #smaps_hat,smaps64_hat = netS(x_hat1,x_hat64) - if epoch>4 or (epoch<4 and ganIterations%10 ~= 9): + if epoch>0: with torch.no_grad(): smaps,smaps64 = netS(target,target_256) L_img_ = 0.33*criterionCAE(x_hat64, target_256) #+ 0.5*criterionCAE(smaps_hat, smaps)