Two-stage-Knowledge-For-Multiple-Adverse-Weather-Removal
Two-stage-Knowledge-For-Multiple-Adverse-Weather-Removal copied to clipboard
fingerk28
[CVPR 2022] Learning Multiple Adverse Weather Removal via Two-stage Knowledge Learning and Multi-contrastive Regularization: Toward a Unified Model
[Setting2] [Snow100k, Raindrop, Rainfog] Download link:https://github.com/fingerk28
I am looking forward to your reply. Thank you very much.
In the first picture ,is the position of LGPiexl and LTPiexl written backwards
Hi, Thank you for your nice work. I want to know how you calculate the metrics for reporting. Here are some results using your code and your model weights. 1)...
In your paper you mention: "At the training stage, we sample 5000 images from "OTS", "Rain 1400", and "CSD" as three individual training sets, respectively.". Could you specify which images...
pBar = tqdm(train_loader, desc='Training') for target_images, input_images in pBar: if target_images is None: continue Every time it runs， target_images and input_images is None.  But my data loader is functioning...
Hi @fingerk28 , thanks for your amazing work! I'm interesting about Setting2, i.e. Snow100k, Raindrop, Rainfog. However, I can't find corresponding test set of [Rainfog](https://github.com/liruoteng/HeavyRainRemoval). So I would like to...
Why the contrastive regularization in the paper is different from the code?
