-
Notifications
You must be signed in to change notification settings - Fork 63
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
3 changed files
with
47 additions
and
16 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -4,6 +4,10 @@ | |
|
||
[![arXiv](https://img.shields.io/badge/arXiv-Paper-<COLOR>.svg)](https://arxiv.org/abs/2209.11345) | ||
![visitors](https://visitor-badge.glitch.me/badge?page_id=mv-lab/swin2sr) | ||
[ <a href="https://colab.research.google.com/drive/1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1?usp=sharing"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="google colab logo"></a>](https://colab.research.google.com/drive/1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1?usp=sharing) | ||
[![Hugging Face Spaces](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Spaces-blue)](https://huggingface.co/spaces/jjourney1125/swin2sr) | ||
[ <a href="https://www.kaggle.com/code/jesucristo/super-resolution-demo-swin2sr-official/"><img src="https://upload.wikimedia.org/wikipedia/commons/7/7c/Kaggle_logo.png?20140912155123" alt="kaggle logo" width=50></a>](https://www.kaggle.com/code/jesucristo/super-resolution-demo-swin2sr-official/) | ||
|
||
|
||
<br> | ||
|
||
|
@@ -20,15 +24,16 @@ MegaStudyEdu, South Korea | |
|
||
**News 🚀🚀** | ||
|
||
- [09/2022] Ongoing website and multiple demos creation. Feel free to contact us. Paper will be presented at AIM, ECCV 2022. | ||
- [10/2022] Demos on Kaggle, Collab and Huggingface Spaces 🤗 are ready! | ||
- [09/2022] Ongoing website and multiple demos creation. Feel free to contact us. Paper will be presented at the [Advances in Image Manipulation (AIM) workshop](https://data.vision.ee.ethz.ch/cvl/aim22/), ECCV 2022, Tel Aviv. | ||
|
||
------------------ | ||
|
||
This is the official repository and PyTorch implementation of Swin2SR. We provide the supplementary material, code, pretrained models and demos. Swin2SR represents a possible improvement of the famous [SwinIR](https://github.com/JingyunLiang/SwinIR/) by [Jingyun Liang](https://jingyunliang.github.io/) (kudos for such an amazing contribution ✋). Our model achieves state-of-the-art performance in: | ||
|
||
- classical, lighweight and real-world image super-resolution (SR) | ||
- color JPEG compression artifact reduction | ||
- compressed input super-resolution: top solution at the "[AIM 2022 Challenge on Super-Resolution of Compressed Image and Video](https://codalab.lisn.upsaclay.fr/competitions/5076)" | ||
- compressed input super-resolution: top solution at the "[AIM 2022 Challenge on Super-Resolution of Compressed Image and Video](https://codalab.lisn.upsaclay.fr/competitions/5076)" organized by [Ren Yang](https://scholar.google.de/citations?hl=en&user=3NgkOp0AAAAJ) and [Radu Timofte](https://scholar.google.de/citations?user=u3MwH5kAAAAJ&hl=en) | ||
|
||
|
||
<p align="center"> | ||
|
@@ -56,6 +61,15 @@ In this paper, we explore the novel Swin Transformer V2, to improve SwinIR for i | |
The training code is at [KAIR](https://github.com/cszn/KAIR/). We follow the same training setup as [SwinIR](https://github.com/JingyunLiang/SwinIR/) by [Jingyun Liang](https://jingyunliang.github.io/). We are working on KAIR integration 👀 | ||
More details about the training setup in our [paper](https://arxiv.org/abs/2209.11345). | ||
|
||
|
||
<details> | ||
<summary><b>Why moving to Swin Transformer V2 ??</b></summary> | ||
<img src="media/paper-why.png " alt="paper swinv2" width="800" border="0"> | ||
Especially in the case of lightweight super-resolution, we noticed how our model convergence was approximately x2 faster using the same experimental setup as SwinIR. We provide the details in the paper Section 3 and 4.2 | ||
</details> | ||
|
||
<br> | ||
|
||
Please check our **[demos](#demos) ready to run** 🚀 | ||
|
||
------ | ||
|
@@ -64,20 +78,20 @@ Please check our **[demos](#demos) ready to run** 🚀 | |
|
||
We achieved state-of-the-art performance on classical, lightweight and real-world image Super-Resolution (SR), JPEG compression artifact reduction, and compressed input super-resolution. We use mainly the DIV2K Dataset and Flickr2K datasets for training, and for testing: RealSRSet, 5images/Classic5/Set5, Set14, BSD100, Urban100 and Manga109 | ||
|
||
🌎 **[All visual results of SwinIR can be downloaded here](https://github.com/mv-lab/swin2sr/releases)**. We also provide links to download the original datasets. | ||
🌎 **[All visual results of Swin2SR can be downloaded here](https://github.com/mv-lab/swin2sr/releases)**. We also provide links to download the original datasets. | ||
More details in our [paper](https://arxiv.org/abs/2209.11345). | ||
|
||
<br> | ||
|
||
<img src="media/chain-sample.png " alt="swin2sr makima demo" width="530" border="0"> | ||
<img src="media/gojou-sample.png " alt="swin2sr makima demo" width="530" border="0"> | ||
|
||
|
||
|Compressed inputs | Swin2SR output| | ||
| :--- | :---: | | ||
| <img src="media/frog_0.png" alt="frog_in" width="250" border="0"> | <img src="media/frog_1.png" alt="frog_swin2sr" width="250" border="0"> | | ||
| <img src="media/0814_0.png" alt="div2k_in" width="250" border="0"> | <img src="media/0814_1.png" alt="div2k_swin2sr" width="250" border="0"> | | ||
| <img src="media/buildings_0.png" alt="buildings_in" width="250" border="0"> | <img src="media/buildings_1.png" alt="buildings_swin2sr" width="250" border="0"> | | ||
| <img src="media/frog_0.png" alt="frog_in" width="300" border="0"> | <img src="media/frog_1.png" alt="frog_swin2sr" width="300" border="0"> | | ||
| <img src="media/0814_0.png" alt="div2k_in" width="300" border="0"> | <img src="media/0814_1.png" alt="div2k_swin2sr" width="300" border="0"> | | ||
| <img src="media/buildings_0.png" alt="buildings_in" width="300" border="0"> | <img src="media/buildings_1.png" alt="buildings_swin2sr" width="300" border="0"> | | ||
|
||
<img src="media/chain-sample.png " alt="swin2sr makima demo" width="620" border="0"> | ||
<img src="media/gojou-sample.png " alt="swin2sr makima demo" width="620" border="0"> | ||
|
||
|
||
<br> | ||
|
||
|
@@ -89,7 +103,7 @@ More details in our [paper](https://arxiv.org/abs/2209.11345). | |
|
||
1. create a folder `inputs` and put there the input images. The model expects low-quality and low-resolution JPEG compressed images. | ||
|
||
2. select `--scale` standard is 4, this means we will increase the resolution of the image x4 times. For example for a 1MP image (1000x1000) we will upscale it to near 4K (4000x4000). We also allow scale 2 (x2 upscaling), for further resolutions please contact us and we will provide the models. | ||
2. select `--scale` standard is 4, this means we will increase the resolution of the image x4 times. For example for a 1MP image (1000x1000) we will upscale it to near 4K (4000x4000). | ||
|
||
3. run our model using `main_test_swin2sr.py` and `--save_img_only`. The pre-trained models are included in [our repo](https://github.com/mv-lab/swin2sr), you can download them from [here](https://github.com/mv-lab/swin2sr/releases) or check the repo [releases](https://github.com/mv-lab/swin2sr/releases). It is important to select the proper `--task`, by default we do compressed input super-resolution `compressed_s`. | ||
|
||
|
@@ -106,14 +120,29 @@ to **reproduce results**, calculate metrics and further evaluation, please check | |
|
||
## Demos | ||
|
||
🔥 🚀 ✅ **[Kaggle kernel demo](https://www.kaggle.com/code/jinssaa/swin2sr-dev/) ready to run!** easy to follow includes testing for multiple SR applications. | ||
|
||
[Super-Resolution Demo Swin2SR Official](https://www.kaggle.com/code/jesucristo/super-resolution-demo-swin2sr-official/) | ||
🔥 🚀 ✅ **[Kaggle kernel demo](https://www.kaggle.com/code/jesucristo/super-resolution-demo-swin2sr-official/) ready to run!** easy to follow includes testing for multiple SR applications. | ||
|
||
<details> | ||
<summary>Clicke here to see how the Kaggle demo looks like</summary> | ||
<p align="center"> | ||
<img src="media/kaggle-demo.png " alt="kaggle demo" width="800" border="0"> | ||
</p> | ||
</details> | ||
|
||
<br> | ||
|
||
[Super-Resolution Demo Swin2SR Official](https://www.kaggle.com/code/jesucristo/super-resolution-demo-swin2sr-official/) is also available in [<a href="https://colab.research.google.com/drive/1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1?usp=sharing">Google Colab <img src="https://colab.research.google.com/assets/colab-badge.svg" alt="google colab logo"></a>](https://colab.research.google.com/drive/1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1?usp=sharing) | ||
|
||
We also have an **interactive demo, no login required!** in [Huggingface Spaces 🤗](https://huggingface.co/spaces/jjourney1125/swin2sr) just click and upload images. | ||
|
||
|
||
<img src="media/hug-demo.png" alt="swin2sr huggingface demo" width="800" border="0"> | ||
|
||
We are working on more interactive demos 👀 Contact us if you have ideas! | ||
|
||
|
||
|
||
---- | ||
<br> | ||
|
||
|
@@ -144,6 +173,7 @@ We follow the same evaluation setup as [SwinIR](https://github.com/JingyunLiang/ | |
### ClassicalSR | ||
``` | ||
python main_test_swin2sr.py --task classical_sr --scale 2 --training_patch_size 64 --model_path model_zoo/swin2sr/Swin2SR_ClassicalSR_X2_64.pth --folder_lq testsets/Set5/LR_bicubic/X2 --folder_gt testsets/Set5/HR | ||
python main_test_swin2sr.py --task classical_sr --scale 4 --training_patch_size 64 --model_path model_zoo/swin2sr/Swin2SR_ClassicalSR_X4_64.pth --folder_lq testsets/Set5/LR_bicubic/X4 --folder_gt testsets/Set5/HR | ||
``` | ||
|
||
|
@@ -177,11 +207,12 @@ python main_test_swin2sr.py --task jpeg_car --jpeg 40 --model_path model_zoo/swi | |
|
||
**[SwinIR: Image Restoration Using Swin Transformer](https://arxiv.org/abs/2108.10257) by Liang et al, ICCVW 2021.** | ||
|
||
[Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Liu et al, CVPR 2022. | ||
**[AISP: AI Image Signal Processing](https://github.com/mv-lab/AISP) by Marcos Conde, Radu Timofte and collaborators, 2022.** | ||
|
||
[AIM 2022 Challenge on Super-Resolution of Compressed Image and Video](https://arxiv.org/abs/2208.11184) organized by Ren Yang. | ||
|
||
**[AISP: AI Image Signal Processing](https://github.com/mv-lab/AISP) by Marcos Conde, Radu Timofte and collaborators, 2022.** | ||
[Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Liu et al, CVPR 2022. | ||
|
||
|
||
----------------- | ||
|
||
|
@@ -208,4 +239,4 @@ This project is released under the Apache 2.0 license. The codes are heavily bas | |
|
||
## Contact | ||
|
||
Marcos Conde ([email protected]) and Ui-Jin Choi are the contact persons. Please add in the email subject "swin2sr". | ||
Marcos Conde ([email protected]) and Ui-Jin Choi ( [email protected]) are the contact persons. Please add in the email subject "swin2sr". |
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.