-
-
Notifications
You must be signed in to change notification settings - Fork 108
/
Copy pathREADME.md
98 lines (69 loc) · 21.8 KB
/
README.md
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
## Introduction
<a href="http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net">Official Repo</a>
<a href="https://github.com/SegmentationBLWX/sssegmentation/blob/main/ssseg/modules/models/backbones/unet.py">Code Snippet</a>
<details>
<summary align="left"><a href="https://arxiv.org/pdf/1505.04597.pdf">UNet (MICCAI'2016/Nat. Methods'2019)</a></summary>
```latex
@inproceedings{ronneberger2015u,
title={U-net: Convolutional networks for biomedical image segmentation},
author={Ronneberger, Olaf and Fischer, Philipp and Brox, Thomas},
booktitle={International Conference on Medical image computing and computer-assisted intervention},
pages={234--241},
year={2015},
organization={Springer}
}
```
</details>
## Results (Cross Entropy)
#### HRF
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.08% | 79.78% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_hrf.log) |
| PSPNet | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.25% | 80.11% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_hrf.log) |
| DeepLabV3 | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.40% | 80.40% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_hrf.log) |
#### DRIVE
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_drive.log) |
| PSPNet | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_drive.log) |
| DeepLabV3 | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_drive.log) |
#### STARE
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.80% | 81.06% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_stare.log) |
| PSPNet | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.94% | 81.35% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_stare.log) |
| DeepLabV3 | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.91% | 81.28% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_stare.log) |
#### CHASE DB1
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.52% | 80.36% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_chasedb1.log) |
| PSPNet | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.57% | 80.48% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_chasedb1.log) |
| DeepLabV3 | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.60% | 80.53% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_chasedb1.log) |
## Results (Cross Entropy + Dice)
#### HRF
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_cedice_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_hrf.log) |
| PSPNet | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_cedice_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_hrf.log) |
| DeepLabV3 | - | UNet-S5-D16 | 256x256 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_cedice_hrf.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_hrf.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_hrf.log) |
#### DRIVE
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_cedice_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_drive.log) |
| PSPNet | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_cedice_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_drive.log) |
| DeepLabV3 | - | UNet-S5-D16 | 64x64 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_cedice_drive.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_drive.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_drive.log) |
#### STARE
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_cedice_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_stare.log) |
| PSPNet | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_cedice_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_stare.log) |
| DeepLabV3 | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | | | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_cedice_stare.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_stare.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_stare.log) |
#### CHASE DB1
| Segmentor | Pretrain | Backbone | Crop Size | Schedule | Train/Eval Set | mDice | Dice | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| FCN | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.59% | 80.50% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/fcn_unets5os16_cedice_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/fcn_unets5os16_cedice_chasedb1.log) |
| PSPNet | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.51% | 80.37% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/pspnet_unets5os16_cedice_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/pspnet_unets5os16_cedice_chasedb1.log) |
| DeepLabV3 | - | UNet-S5-D16 | 128x128 | LR/POLICY/BS/EPOCH: 0.01/poly/16/1 | train/val | 89.56% | 80.44% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/unet/deeplabv3_unets5os16_cedice_chasedb1.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_chasedb1.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_unet/deeplabv3_unets5os16_cedice_chasedb1.log) |
## More
You can also download the model weights from following sources:
- BaiduNetdisk: https://pan.baidu.com/s/1gD-NJJWOtaHCtB0qHE79rA with access code **s757**