KAIR
KAIR copied to clipboard
cszn
Image Restoration Toolbox (PyTorch). Training and testing codes for DPIR, USRNet, DnCNN, FFDNet, SRMD, DPSR, BSRGAN, SwinIR
Hello Everyone, Could anyone share the modified json file to finetune the super resolution for real world images ( for example json : options/swinir/train_swinir_sr_realworld_psnr.json ) It would be also great...
Goal: Deblur images with a paired dataset sharp/blurred using SwinIR? In your repository you provide a dataset type "Plain" located at data/dataset_plain.py Is this the right dataset loader to use...
Hi there, I was attempting to reproduce the ffdnet_gray.pth model in the model_zoo. After checking the training option, I found the training dataset setting is somehow unclear to me: which...
Hi, I was checking training details on the [BSRGAN](https://arxiv.org/pdf/2103.14006.pdf) paper, and comparing them to `options/train_bsrgan_x4_gan.json`, I found a few differences which might be important for reproduction and research. I'm working...
In [model_gan.py](https://github.com/cszn/KAIR/blob/master/models/model_gan.py#L222) we have: ```python if self.opt['train']['gan_type'] in ['gan', 'lsgan', 'wgan', 'softplusgan']: pred_g_fake = self.netD(self.E) D_loss = self.D_lossfn_weight * self.D_lossfn(pred_g_fake, True) ``` But when D & G have converged, `pred_g_fake`...
Hello, I am training USRNet for a deblurring task and now it seems to have a nice result after tring some noise level. But I have a quetion when I...
https://github.com/cszn/KAIR/blob/master/main_test_face_enhancement.py I want to run main_test_face_enhancement.py, but I can't get RetinaFace-R50.pth and GPEN-512.pth. ================================================== How to run: step 1: Download model and model and put them into `model_zoo`. RetinaFace-R50.pth: https://public-vigen-video.oss-cn-shanghai.aliyuncs.com/robin/models/RetinaFace-R50.pth...
I have the DPSR training running but it does not appear to be fully utilising the GPU which spikes in a saw too profile as below:  I've tried adjusting...
https://github.com/zzr-idam/Under-Display-Camera-UAV
I tried srmd-ncnn-vulkan and python main_test_srmd.py, and found that the conversion results of SRMD are both very bad and very different from the demo. copy these image to KAIR-master\testsets\set5 ...
