Skip to content

Commit 7f4fa79

Browse files
ckkelvinchaninnerlee
ckkelvinchan
and
innerlee
authored
[Doc] Update model zoo for documents (open-mmlab#400)
* Update model zoo for documents * Collapse instructions * bold them * hr br * cool Co-authored-by: lizz <lizz@sensetime.com>
1 parent 3b3cbf2 commit 7f4fa79

File tree

25 files changed

+194
-149
lines changed

25 files changed

+194
-149
lines changed

‎configs/inpainting/deepfillv1/README.md

+9-6
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,10 @@
1-
# Generative Image Inpainting with Contextual Attention
2-
3-
## Introduction
1+
# DeepFillv1 (CVPR'2018)
42

53
<!-- [ALGORITHM] -->
64

5+
<details>
6+
<summary align="right">DeepFillv1 (CVPR'2018)</summary>
7+
78
```bibtex
89
@inproceedings{yu2018generative,
910
title={Generative image inpainting with contextual attention},
@@ -14,15 +15,17 @@
1415
}
1516
```
1617

17-
## Results and models
18+
</details>
19+
20+
<br/>
1821

19-
### Places365-Challenge
22+
**Places365-Challenge**
2023

2124
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2225
| :--------------------------------------------------------------------------: | :---------: | :--------: | :---------: | :-----------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
2326
| [DeepFillv1](/configs/inpainting/deepfillv1/deepfillv1_256x256_8x2_places.py) | square bbox | 256x256 | 3500k | Places365-val | 11.019 | 23.429 | 0.862 | [model](https://download.openmmlab.com/mmediting/inpainting/deepfillv1/deepfillv1_256x256_8x2_places_20200619-c00a0e21.pth) \| [log](https://download.openmmlab.com/mmediting/inpainting/deepfillv1/deepfillv1_256x256_8x2_places_20200619-c00a0e21.log.json) |
2427

25-
### CelebA-HQ
28+
**CelebA-HQ**
2629

2730
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2831
| :--------------------------------------------------------------------------: | :---------: | :--------: | :---------: | :--------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/inpainting/deepfillv2/README.md

+8-6
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Free-form Image Inpainting with Gated Convolution
2-
3-
## Introduction
1+
# DeepFillv2 (CVPR'2019)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
<summary align="right">DeepFillv2 (CVPR'2019)</summary>
66

77
```bibtex
88
@inproceedings{yu2019free,
@@ -14,15 +14,17 @@
1414
}
1515
```
1616

17-
## Results and models
17+
</details>
18+
19+
<br/>
1820

19-
### Places365-Challenge
21+
**Places365-Challenge**
2022

2123
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2224
| :--------------------------------------------------------------------------: | :-------: | :--------: | :---------: | :-----------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
2325
| [DeepFillv2](/configs/inpainting/deepfillv2/deepfillv2_256x256_8x2_places.py) | free-form | 256x256 | 100k | Places365-val | 8.635 | 22.398 | 0.815 | [model](https://download.openmmlab.com/mmediting/inpainting/deepfillv2/deepfillv2_256x256_8x2_places_20200619-10d15793.pth) \| [log](https://download.openmmlab.com/mmediting/inpainting/deepfillv2/deepfillv2_256x256_8x2_places_20200619-10d15793.log.json) |
2426

25-
### CelebA-HQ
27+
**CelebA-HQ**
2628

2729
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2830
| :--------------------------------------------------------------------------: | :-------: | :--------: | :---------: | :--------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/inpainting/global_local/README.md

+9-7
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Globally and Locally Consistent Image Completion
2-
3-
## Introduction
1+
# Global&Local (ToG'2017)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
<summary align="right">Global&Local (ToG'2017)</summary>
66

77
```bibtex
88
@article{iizuka2017globally,
@@ -17,17 +17,19 @@
1717
}
1818
```
1919

20-
*Note that we do not apply the post-processing module in Global&Local for a fair comparison with current deep inpainting methods.*
20+
</details>
2121

22-
## Results and models
22+
<br/>
23+
24+
*Note that we do not apply the post-processing module in Global&Local for a fair comparison with current deep inpainting methods.*
2325

24-
### Places365-Challenge
26+
**Places365-Challenge**
2527

2628
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2729
| :-----------------------------------------------------------------------: | :---------: | :--------: | :---------: | :-----------: | :------: | :----: | :---: | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
2830
| [Global&Local](/configs/inpainting/global_local/gl_256x256_8x12_places.py) | square bbox | 256x256 | 500k | Places365-val | 11.164 | 23.152 | 0.862 | [model](https://download.openmmlab.com/mmediting/inpainting/global_local/gl_256x256_8x12_places_20200619-52a040a8.pth) \| [log](https://download.openmmlab.com/mmediting/inpainting/global_local/gl_256x256_8x12_places_20200619-52a040a8.log.json) |
2931

30-
### CelebA-HQ
32+
**CelebA-HQ**
3133

3234
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
3335
| :-----------------------------------------------------------------------: | :---------: | :--------: | :---------: | :--------: | :------: | :----: | :---: | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/inpainting/partial_conv/README.md

+9-6
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,10 @@
1-
# Image inpainting for Irregular Holes using Partial Convolutions
2-
3-
## Introduction
1+
# PConv (ECCV'2018)
42

53
<!-- [ALGORITHM] -->
64

5+
<details>
6+
<summary align="right">PConv (ECCV'2018)</summary>
7+
78
```bibtex
89
@inproceedings{liu2018image,
910
title={Image inpainting for irregular holes using partial convolutions},
@@ -14,15 +15,17 @@
1415
}
1516
```
1617

17-
## Results and models
18+
</details>
19+
20+
<br/>
1821

19-
### Places365-Challenge
22+
**Places365-Challenge**
2023

2124
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2225
| :-------------------------------------------------------------------------: | :-------: | :--------: | :---------: | :-----------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
2326
| [PConv](/configs/inpainting/partial_conv/pconv_256x256_stage2_4x2_places.py) | free-form | 256x256 | 500k | Places365-val | 8.776 | 22.762 | 0.801 | [model](https://download.openmmlab.com/mmediting/inpainting/pconv/pconv_256x256_stage2_4x2_places_20200619-1ffed0e8.pth) \| [log](https://download.openmmlab.com/mmediting/inpainting/pconv/pconv_256x256_stage2_4x2_places_20200619-1ffed0e8.log.json) |
2427

25-
### CelebA-HQ
28+
**CelebA-HQ**
2629

2730
| Method | Mask Type | Resolution | Train Iters | Test Set | l1 error | PSNR | SSIM | Download |
2831
| :-------------------------------------------------------------------------: | :-------: | :--------: | :---------: | :--------: | :------: | :----: | :---: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/mattors/dim/README.md

+6-4
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Deep Image Matting
2-
3-
## Introduction
1+
# DIM (CVPR'2017)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
<summary align="right">DIM (CVPR'2017)</summary>
66

77
```bibtex
88
@inproceedings{xu2017deep,
@@ -14,7 +14,9 @@
1414
}
1515
```
1616

17-
## Results and Models
17+
</details>
18+
19+
<br/>
1820

1921
| Method | SAD | MSE | GRAD | CONN | Download |
2022
| :------------------------------------------------------------------------: | :------: | :-------: | :------: | :------: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/mattors/gca/README.md

+7-5
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Natural Image Matting via Guided Contextual Attention
2-
3-
## Introduction
1+
# GCA (AAAI'2020)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
<summary align="right">GCA (AAAI'2020)</summary>
66

77
```bibtex
88
@inproceedings{li2020natural,
@@ -13,7 +13,9 @@
1313
}
1414
```
1515

16-
## Results and Models
16+
</details>
17+
18+
<br/>
1719

1820
| Method | SAD | MSE | GRAD | CONN | Download |
1921
| :--------------------------------------------------------------------: | :-------: | :--------: | :-------: | :-------: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
@@ -22,7 +24,7 @@
2224
| [baseline (our)](/configs/mattors/gca/baseline_r34_4x10_200k_comp1k.py) | 36.50 | 0.0090 | 17.40 | 34.33 | [model](https://download.openmmlab.com/mmediting/mattors/gca/baseline_r34_4x10_200k_comp1k_SAD-36.50_20200614_105701-95be1750.pth) \| [log](https://download.openmmlab.com/mmediting/mattors/gca/baseline_r34_4x10_200k_comp1k_20200614_105701.log.json) |
2325
| [GCA (our)](/configs/mattors/gca/gca_r34_4x10_200k_comp1k.py) | **34.77** | **0.0080** | **16.33** | **32.20** | [model](https://download.openmmlab.com/mmediting/mattors/gca/gca_r34_4x10_200k_comp1k_SAD-34.77_20200604_213848-4369bea0.pth) \| [log](https://download.openmmlab.com/mmediting/mattors/gca/gca_r34_4x10_200k_comp1k_20200604_213848.log.json) |
2426

25-
### More results
27+
**More results**
2628

2729
| Method | SAD | MSE | GRAD | CONN | Download |
2830
| :-----------------------------------------------------------------------------------------: | :---: | :----: | :---: | :---: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/mattors/indexnet/README.md

+7-5
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,8 @@
1-
# Indices Matter: Learning to Index for Deep Image Matting
2-
3-
## Introduction
1+
# IndexNet (ICCV'2019)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
<summary align="right">IndexNet (ICCV'2019)</summary>
66

77
```bibtex
88
@inproceedings{hao2019indexnet,
@@ -13,7 +13,9 @@
1313
}
1414
```
1515

16-
## Results and Models
16+
</details>
17+
18+
<br/>
1719

1820
| Method | SAD | MSE | GRAD | CONN | Download |
1921
| :--------------------------------------------------------------------------: | :------: | :-------: | :------: | :------: | :-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
@@ -22,7 +24,7 @@
2224

2325
> The performance of training (best performance) with different random seeds diverges in a large range. You may need to run several experiments for each setting to obtain the above performance.
2426
25-
### More result
27+
**More result**
2628

2729
| Method | SAD | MSE | GRAD | CONN | Download |
2830
| :-----------------------------------------------------------------------------------------------: | :---: | :---: | :---: | :---: | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |

‎configs/restorers/basicvsr/README.md

+8-7
Original file line numberDiff line numberDiff line change
@@ -1,10 +1,11 @@
1-
# BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond
2-
3-
## Introduction
1+
# BasicVSR (CVPR'2021)
42

53
<!-- [ALGORITHM] -->
64

7-
```
5+
<details>
6+
<summary align="right">BasicVSR (CVPR'2021)</summary>
7+
8+
```bibtex
89
@InProceedings{chan2021basicvsr,
910
author = {Chan, Kelvin CK and Wang, Xintao and Yu, Ke and Dong, Chao and Loy, Chen Change},
1011
title = {BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond},
@@ -13,13 +14,13 @@
1314
}
1415
```
1516

16-
## Results and Models
17+
</details>
1718

18-
Evaluated on RGB channels for REDS4 and Y channel for others. The metrics are `PSNR`/`SSIM`.
19+
<br/>
1920

21+
Evaluated on RGB channels for REDS4 and Y channel for others. The metrics are `PSNR` / `SSIM` .
2022
The pretrained weights of SPyNet can be found [here](https://download.openmmlab.com/mmediting/restorers/basicvsr/spynet_20210409-c6c1bd09.pth).
2123

22-
2324
| Method | REDS4 (BIx4)<br>PSNR/SSIM (RGB) | Vimeo-90K-T (BIx4)<br>PSNR/SSIM (Y) | Vid4 (BIx4)<br>PSNR/SSIM (Y) | UDM10 (BDx4)<br>PSNR/SSIM (Y) | Vimeo-90K-T (BDx4)<br>PSNR/SSIM (Y) | Vid4 (BDx4)<br>PSNR/SSIM (Y) | Download |
2425
|:------------------------------------------------------------------------------------------------------------------------------:|:-------------------------------:|:-----------------------------------:|:----------------------------:|:-----------------------------:|:-----------------------------------:|:----------------------------:|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
2526
| [basicvsr_reds4](https://github.com/open-mmlab/mmediting/blob/master/configs/restorers/basicvsr/basicvsr_reds4.py) | **31.4170/0.8909** | 36.2848/0.9395 | 27.2694/0.8318 | 33.4478/0.9306 | 34.4700/0.9286 | 24.4541/0.7455 | [model](https://download.openmmlab.com/mmediting/restorers/basicvsr/basicvsr_reds4_20120409-0e599677.pth) \| [log](https://download.openmmlab.com/mmediting/restorers/basicvsr/basicvsr_reds4_20210409_092646.log.json) |

‎configs/restorers/dic/README.md

+8-6
Original file line numberDiff line numberDiff line change
@@ -1,8 +1,9 @@
1-
# Learning Texture Transformer Network for Image Super-Resolution
2-
3-
## Introduction
1+
# DIC (CVPR'2020)
42

53
<!-- [ALGORITHM] -->
4+
<details>
5+
6+
<summary align="right">DIC (CVPR'2020)</summary>
67

78
```bibtex
89
@inproceedings{ma2020deep,
@@ -14,11 +15,12 @@
1415
}
1516
```
1617

17-
## Results
18+
</details>
1819

19-
Evaluated on RGB channels, `scale` pixels in each border are cropped before evaluation.
20+
<br/>
2021

21-
The metrics are `PSNR / SSIM`.
22+
Evaluated on RGB channels, `scale` pixels in each border are cropped before evaluation.
23+
The metrics are `PSNR / SSIM` .
2224

2325
In the log data of `dic_gan_x8c48b6_g4_150k_CelebAHQ`, DICGAN is verified on the first 9 pictures of the test set of CelebA-HQ, so `PSNR/SSIM` shown in the follow table is different from the log data.
2426

0 commit comments

Comments
 (0)