-
-
Notifications
You must be signed in to change notification settings - Fork 110
Expand file tree
/
Copy pathREADME.md
More file actions
51 lines (34 loc) · 7.01 KB
/
README.md
File metadata and controls
51 lines (34 loc) · 7.01 KB
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
## Introduction
<a href="https://github.com/facebookresearch/detectron2/tree/master/projects/PointRend">Official Repo</a>
<a href="https://github.com/SegmentationBLWX/sssegmentation/blob/main/ssseg/modules/models/segmentors/pointrend/pointrend.py">Code Snippet</a>
<details>
<summary align="left"><a href="https://arxiv.org/pdf/1912.08193.pdf">PointRend (CVPR'2020)</a></summary>
```latex
@inproceedings{kirillov2020pointrend,
title={Pointrend: Image segmentation as rendering},
author={Kirillov, Alexander and Wu, Yuxin and He, Kaiming and Girshick, Ross},
booktitle={Proceedings of the IEEE/CVF conference on computer vision and pattern recognition},
pages={9799--9808},
year={2020}
}
```
</details>
## Results
#### PASCAL VOC
| Backbone | Pretrain | Crop Size | Schedule | Train/Eval Set | mIoU | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| R-50-D32 | ImageNet-1k-224x224 | 512x512 | LR/POLICY/BS/EPOCH: 0.01/poly/16/60 | trainaug/val | 69.84% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet50os32_voc.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_voc.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_voc.log) |
| R-101-D32 | ImageNet-1k-224x224 | 512x512 | LR/POLICY/BS/EPOCH: 0.01/poly/16/60 | trainaug/val | 72.31% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet101os32_voc.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_voc.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_voc.log) |
#### ADE20k
| Backbone | Pretrain | Crop Size | Schedule | Train/Eval Set | mIoU | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| R-50-D32 | ImageNet-1k-224x224 | 512x512 | LR/POLICY/BS/EPOCH: 0.01/poly/16/130 | train/val | 37.80% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet50os32_ade20k.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_ade20k.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_ade20k.log) |
| R-101-D32 | ImageNet-1k-224x224 | 512x512 | LR/POLICY/BS/EPOCH: 0.01/poly/16/130 | train/val | 40.26% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet101os32_ade20k.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_ade20k.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_ade20k.log) |
#### CityScapes
| Backbone | Pretrain | Crop Size | Schedule | Train/Eval Set | mIoU | Download |
| :-: | :-: | :-: | :-: | :-: | :-: | :-: |
| R-50-D32 | ImageNet-1k-224x224 | 512x1024 | LR/POLICY/BS/EPOCH: 0.01/poly/8/220 | train/val | 76.89% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet50os32_cityscapes.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_cityscapes.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet50os32_cityscapes.log) |
| R-101-D32 | ImageNet-1k-224x224 | 512x1024 | LR/POLICY/BS/EPOCH: 0.01/poly/8/220 | train/val | 78.80% | [cfg](https://raw.githubusercontent.com/SegmentationBLWX/sssegmentation/main/ssseg/configs/pointrend/pointrend_resnet101os32_cityscapes.py) | [model](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_cityscapes.pth) | [log](https://github.com/SegmentationBLWX/modelstore/releases/download/ssseg_pointrend/pointrend_resnet101os32_cityscapes.log) |
## More
You can also download the model weights from following sources:
- BaiduNetdisk: https://pan.baidu.com/s/1gD-NJJWOtaHCtB0qHE79rA with access code **s757**