admin/Seg_Data_Server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

first commit

Browse Source
This commit is contained in:
admin
2026-05-20 15:05:35 +08:00
commit ac09b26253
2048 changed files with 189478 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

68
Seg_All_In_One_MMSeg/configs/psanet/README.md Normal file
View File

@@ -0,0 +1,68 @@
# PSANet
> [PSANet: Point-wise Spatial Attention Network for Scene Parsing](https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf)
## Introduction
<!-- [ALGORITHM] -->
<a href="https://github.com/hszhao/PSANet">Official Repo</a>
<a href="https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18">Code Snippet</a>
## Abstract
<!-- [ABSTRACT] -->
We notice information flow in convolutional neural networksis restricted inside local neighborhood regions due to the physical de-sign of convolutional filters, which limits the overall understanding ofcomplex scenes. In this paper, we propose thepoint-wise spatial atten-tion network(PSANet) to relax the local neighborhood constraint. Eachposition on the feature map is connected to all the other ones througha self-adaptively learned attention mask. Moreover, information propa-gation in bi-direction for scene parsing is enabled. Information at otherpositions can be collected to help the prediction of the current positionand vice versa, information at the current position can be distributedto assist the prediction of other ones. Our proposed approach achievestop performance on various competitive scene parsing datasets, includ-ing ADE20K, PASCAL VOC 2012 and Cityscapes, demonstrating itseffectiveness and generality.
<!-- [IMAGE] -->
<div align=center>
<img src="https://user-images.githubusercontent.com/24582831/142902367-0f29e8cb-5ac0-434b-98c4-b2af7c9c2e58.png" width="70%"/>
</div>
## Results and models
### Cityscapes
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
| ------ | -------- | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | ------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| PSANet | R-50-D8 | 512x1024 | 40000 | 7 | 3.17 | V100 | 77.63 | 79.04 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_40k_cityscapes/psanet_r50-d8_512x1024_40k_cityscapes_20200606_103117-99fac37c.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_40k_cityscapes/psanet_r50-d8_512x1024_40k_cityscapes_20200606_103117.log.json) |
| PSANet | R-101-D8 | 512x1024 | 40000 | 10.5 | 2.20 | V100 | 79.14 | 80.19 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_40k_cityscapes/psanet_r101-d8_512x1024_40k_cityscapes_20200606_001418-27b9cfa7.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_40k_cityscapes/psanet_r101-d8_512x1024_40k_cityscapes_20200606_001418.log.json) |
| PSANet | R-50-D8 | 769x769 | 40000 | 7.9 | 1.40 | V100 | 77.99 | 79.64 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-769x769.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_40k_cityscapes/psanet_r50-d8_769x769_40k_cityscapes_20200530_033717-d5365506.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_40k_cityscapes/psanet_r50-d8_769x769_40k_cityscapes_20200530_033717.log.json) |
| PSANet | R-101-D8 | 769x769 | 40000 | 11.9 | 0.98 | V100 | 78.43 | 80.26 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-769x769.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_40k_cityscapes/psanet_r101-d8_769x769_40k_cityscapes_20200530_035107-997da1e6.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_40k_cityscapes/psanet_r101-d8_769x769_40k_cityscapes_20200530_035107.log.json) |
| PSANet | R-50-D8 | 512x1024 | 80000 | - | - | V100 | 77.24 | 78.69 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_80k_cityscapes/psanet_r50-d8_512x1024_80k_cityscapes_20200606_161842-ab60a24f.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_80k_cityscapes/psanet_r50-d8_512x1024_80k_cityscapes_20200606_161842.log.json) |
| PSANet | R-101-D8 | 512x1024 | 80000 | - | - | V100 | 79.31 | 80.53 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-512x1024.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_80k_cityscapes/psanet_r101-d8_512x1024_80k_cityscapes_20200606_161823-0f73a169.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_80k_cityscapes/psanet_r101-d8_512x1024_80k_cityscapes_20200606_161823.log.json) |
| PSANet | R-50-D8 | 769x769 | 80000 | - | - | V100 | 79.31 | 80.91 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-769x769.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_80k_cityscapes/psanet_r50-d8_769x769_80k_cityscapes_20200606_225134-fe42f49e.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_80k_cityscapes/psanet_r50-d8_769x769_80k_cityscapes_20200606_225134.log.json) |
| PSANet | R-101-D8 | 769x769 | 80000 | - | - | V100 | 79.69 | 80.89 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-769x769.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_80k_cityscapes/psanet_r101-d8_769x769_80k_cityscapes_20200606_214550-7665827b.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_80k_cityscapes/psanet_r101-d8_769x769_80k_cityscapes_20200606_214550.log.json) |
### ADE20K
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
| ------ | -------- | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | -------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| PSANet | R-50-D8 | 512x512 | 80000 | 9 | 18.91 | V100 | 41.14 | 41.91 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_80k_ade20k/psanet_r50-d8_512x512_80k_ade20k_20200614_144141-835e4b97.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_80k_ade20k/psanet_r50-d8_512x512_80k_ade20k_20200614_144141.log.json) |
| PSANet | R-101-D8 | 512x512 | 80000 | 12.5 | 13.13 | V100 | 43.80 | 44.75 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb4-80k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_80k_ade20k/psanet_r101-d8_512x512_80k_ade20k_20200614_185117-1fab60d4.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_80k_ade20k/psanet_r101-d8_512x512_80k_ade20k_20200614_185117.log.json) |
| PSANet | R-50-D8 | 512x512 | 160000 | - | - | V100 | 41.67 | 42.95 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb4-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_160k_ade20k/psanet_r50-d8_512x512_160k_ade20k_20200615_161258-148077dd.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_160k_ade20k/psanet_r50-d8_512x512_160k_ade20k_20200615_161258.log.json) |
| PSANet | R-101-D8 | 512x512 | 160000 | - | - | V100 | 43.74 | 45.38 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb4-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_160k_ade20k/psanet_r101-d8_512x512_160k_ade20k_20200615_161537-dbfa564c.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_160k_ade20k/psanet_r101-d8_512x512_160k_ade20k_20200615_161537.log.json) |
### Pascal VOC 2012 + Aug
| Method | Backbone | Crop Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
| ------ | -------- | --------- | ------: | -------- | -------------- | ------ | ----: | ------------: | --------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| PSANet | R-50-D8 | 512x512 | 20000 | 6.9 | 18.24 | V100 | 76.39 | 77.34 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb4-20k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_20k_voc12aug/psanet_r50-d8_512x512_20k_voc12aug_20200617_102413-2f1bbaa1.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_20k_voc12aug/psanet_r50-d8_512x512_20k_voc12aug_20200617_102413.log.json) |
| PSANet | R-101-D8 | 512x512 | 20000 | 10.4 | 12.63 | V100 | 77.91 | 79.30 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb4-20k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_20k_voc12aug/psanet_r101-d8_512x512_20k_voc12aug_20200617_110624-946fef11.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_20k_voc12aug/psanet_r101-d8_512x512_20k_voc12aug_20200617_110624.log.json) |
| PSANet | R-50-D8 | 512x512 | 40000 | - | - | V100 | 76.30 | 77.35 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r50-d8_4xb4-40k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_40k_voc12aug/psanet_r50-d8_512x512_40k_voc12aug_20200613_161946-f596afb5.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_40k_voc12aug/psanet_r50-d8_512x512_40k_voc12aug_20200613_161946.log.json) |
| PSANet | R-101-D8 | 512x512 | 40000 | - | - | V100 | 77.73 | 79.05 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/psanet/psanet_r101-d8_4xb4-40k_voc12aug-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_40k_voc12aug/psanet_r101-d8_512x512_40k_voc12aug_20200613_161946-1f560f9e.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_40k_voc12aug/psanet_r101-d8_512x512_40k_voc12aug_20200613_161946.log.json) |
## Citation
```bibtex
@inproceedings{zhao2018psanet,
title={Psanet: Point-wise spatial attention network for scene parsing},
author={Zhao, Hengshuang and Zhang, Yi and Liu, Shu and Shi, Jianping and Change Loy, Chen and Lin, Dahua and Jia, Jiaya},
booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
pages={267--283},
year={2018}
}
```

391
Seg_All_In_One_MMSeg/configs/psanet/metafile.yaml Normal file
View File

@@ -0,0 +1,391 @@
Collections:
- Name: PSANet
License: Apache License 2.0
Metadata:
Training Data:
- Cityscapes
- ADE20K
- Pascal VOC 2012 + Aug
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
README: configs/psanet/README.md
Frameworks:
- PyTorch
Models:
- Name: psanet_r50-d8_4xb2-40k_cityscapes-512x1024
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 77.63
mIoU(ms+flip): 79.04
Config: configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-512x1024.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 7.0
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_40k_cityscapes/psanet_r50-d8_512x1024_40k_cityscapes_20200606_103117-99fac37c.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_40k_cityscapes/psanet_r50-d8_512x1024_40k_cityscapes_20200606_103117.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb2-40k_cityscapes-512x1024
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 79.14
mIoU(ms+flip): 80.19
Config: configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-512x1024.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 10.5
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_40k_cityscapes/psanet_r101-d8_512x1024_40k_cityscapes_20200606_001418-27b9cfa7.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_40k_cityscapes/psanet_r101-d8_512x1024_40k_cityscapes_20200606_001418.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb2-40k_cityscapes-769x769
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 77.99
mIoU(ms+flip): 79.64
Config: configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-769x769.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 7.9
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_40k_cityscapes/psanet_r50-d8_769x769_40k_cityscapes_20200530_033717-d5365506.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_40k_cityscapes/psanet_r50-d8_769x769_40k_cityscapes_20200530_033717.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb2-40k_cityscapes-769x769
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 78.43
mIoU(ms+flip): 80.26
Config: configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-769x769.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 11.9
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_40k_cityscapes/psanet_r101-d8_769x769_40k_cityscapes_20200530_035107-997da1e6.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_40k_cityscapes/psanet_r101-d8_769x769_40k_cityscapes_20200530_035107.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb2-80k_cityscapes-512x1024
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 77.24
mIoU(ms+flip): 78.69
Config: configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-512x1024.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_80k_cityscapes/psanet_r50-d8_512x1024_80k_cityscapes_20200606_161842-ab60a24f.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x1024_80k_cityscapes/psanet_r50-d8_512x1024_80k_cityscapes_20200606_161842.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb2-80k_cityscapes-512x1024
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 79.31
mIoU(ms+flip): 80.53
Config: configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-512x1024.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_80k_cityscapes/psanet_r101-d8_512x1024_80k_cityscapes_20200606_161823-0f73a169.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x1024_80k_cityscapes/psanet_r101-d8_512x1024_80k_cityscapes_20200606_161823.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb2-80k_cityscapes-769x769
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 79.31
mIoU(ms+flip): 80.91
Config: configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-769x769.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_80k_cityscapes/psanet_r50-d8_769x769_80k_cityscapes_20200606_225134-fe42f49e.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_769x769_80k_cityscapes/psanet_r50-d8_769x769_80k_cityscapes_20200606_225134.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb2-80k_cityscapes-769x769
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Cityscapes
Metrics:
mIoU: 79.69
mIoU(ms+flip): 80.89
Config: configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-769x769.py
Metadata:
Training Data: Cityscapes
Batch Size: 8
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_80k_cityscapes/psanet_r101-d8_769x769_80k_cityscapes_20200606_214550-7665827b.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_769x769_80k_cityscapes/psanet_r101-d8_769x769_80k_cityscapes_20200606_214550.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb4-80k_ade20k-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 41.14
mIoU(ms+flip): 41.91
Config: configs/psanet/psanet_r50-d8_4xb4-80k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 9.0
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_80k_ade20k/psanet_r50-d8_512x512_80k_ade20k_20200614_144141-835e4b97.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_80k_ade20k/psanet_r50-d8_512x512_80k_ade20k_20200614_144141.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb4-80k_ade20k-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 43.8
mIoU(ms+flip): 44.75
Config: configs/psanet/psanet_r101-d8_4xb4-80k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 12.5
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_80k_ade20k/psanet_r101-d8_512x512_80k_ade20k_20200614_185117-1fab60d4.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_80k_ade20k/psanet_r101-d8_512x512_80k_ade20k_20200614_185117.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb4-160k_ade20k-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 41.67
mIoU(ms+flip): 42.95
Config: configs/psanet/psanet_r50-d8_4xb4-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_160k_ade20k/psanet_r50-d8_512x512_160k_ade20k_20200615_161258-148077dd.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_160k_ade20k/psanet_r50-d8_512x512_160k_ade20k_20200615_161258.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb4-160k_ade20k-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 43.74
mIoU(ms+flip): 45.38
Config: configs/psanet/psanet_r101-d8_4xb4-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_160k_ade20k/psanet_r101-d8_512x512_160k_ade20k_20200615_161537-dbfa564c.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_160k_ade20k/psanet_r101-d8_512x512_160k_ade20k_20200615_161537.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb4-20k_voc12aug-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Pascal VOC 2012 + Aug
Metrics:
mIoU: 76.39
mIoU(ms+flip): 77.34
Config: configs/psanet/psanet_r50-d8_4xb4-20k_voc12aug-512x512.py
Metadata:
Training Data: Pascal VOC 2012 + Aug
Batch Size: 16
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 6.9
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_20k_voc12aug/psanet_r50-d8_512x512_20k_voc12aug_20200617_102413-2f1bbaa1.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_20k_voc12aug/psanet_r50-d8_512x512_20k_voc12aug_20200617_102413.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb4-20k_voc12aug-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Pascal VOC 2012 + Aug
Metrics:
mIoU: 77.91
mIoU(ms+flip): 79.3
Config: configs/psanet/psanet_r101-d8_4xb4-20k_voc12aug-512x512.py
Metadata:
Training Data: Pascal VOC 2012 + Aug
Batch Size: 16
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Memory (GB): 10.4
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_20k_voc12aug/psanet_r101-d8_512x512_20k_voc12aug_20200617_110624-946fef11.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_20k_voc12aug/psanet_r101-d8_512x512_20k_voc12aug_20200617_110624.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r50-d8_4xb4-40k_voc12aug-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Pascal VOC 2012 + Aug
Metrics:
mIoU: 76.3
mIoU(ms+flip): 77.35
Config: configs/psanet/psanet_r50-d8_4xb4-40k_voc12aug-512x512.py
Metadata:
Training Data: Pascal VOC 2012 + Aug
Batch Size: 16
Architecture:
- R-50-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_40k_voc12aug/psanet_r50-d8_512x512_40k_voc12aug_20200613_161946-f596afb5.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r50-d8_512x512_40k_voc12aug/psanet_r50-d8_512x512_40k_voc12aug_20200613_161946.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch
- Name: psanet_r101-d8_4xb4-40k_voc12aug-512x512
In Collection: PSANet
Results:
Task: Semantic Segmentation
Dataset: Pascal VOC 2012 + Aug
Metrics:
mIoU: 77.73
mIoU(ms+flip): 79.05
Config: configs/psanet/psanet_r101-d8_4xb4-40k_voc12aug-512x512.py
Metadata:
Training Data: Pascal VOC 2012 + Aug
Batch Size: 16
Architecture:
- R-101-D8
- PSANet
Training Resources: 4x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_40k_voc12aug/psanet_r101-d8_512x512_40k_voc12aug_20200613_161946-1f560f9e.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/psanet/psanet_r101-d8_512x512_40k_voc12aug/psanet_r101-d8_512x512_40k_voc12aug_20200613_161946.log.json
Paper:
Title: 'PSANet: Point-wise Spatial Attention Network for Scene Parsing'
URL: https://openaccess.thecvf.com/content_ECCV_2018/papers/Hengshuang_Zhao_PSANet_Point-wise_Spatial_ECCV_2018_paper.pdf
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/decode_heads/psa_head.py#L18
Framework: PyTorch

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-512x1024.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb2-40k_cityscapes-512x1024.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb2-40k_cityscapes-769x769.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb2-40k_cityscapes-769x769.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-512x1024.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb2-80k_cityscapes-512x1024.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb2-80k_cityscapes-769x769.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb2-80k_cityscapes-769x769.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb4-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb4-160k_ade20k-512x512.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb4-20k_voc12aug-512x512.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb4-20k_voc12aug-512x512.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb4-40k_voc12aug-512x512.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb4-40k_voc12aug-512x512.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

2
Seg_All_In_One_MMSeg/configs/psanet/psanet_r101-d8_4xb4-80k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,2 @@
_base_ = './psanet_r50-d8_4xb4-80k_ade20k-512x512.py'
model = dict(pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101))

7
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-512x1024.py Normal file
View File

@@ -0,0 +1,7 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py', '../_base_/datasets/cityscapes.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_40k.py'
]
crop_size = (512, 1024)
data_preprocessor = dict(size=crop_size)
model = dict(data_preprocessor=data_preprocessor)

12
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb2-40k_cityscapes-769x769.py Normal file
View File

@@ -0,0 +1,12 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py',
'../_base_/datasets/cityscapes_769x769.py', '../_base_/default_runtime.py',
'../_base_/schedules/schedule_40k.py'
]
crop_size = (769, 769)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(align_corners=True),
auxiliary_head=dict(align_corners=True),
test_cfg=dict(mode='slide', crop_size=(769, 769), stride=(513, 513)))

7
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-512x1024.py Normal file
View File

@@ -0,0 +1,7 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py', '../_base_/datasets/cityscapes.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
]
crop_size = (512, 1024)
data_preprocessor = dict(size=crop_size)
model = dict(data_preprocessor=data_preprocessor)

12
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb2-80k_cityscapes-769x769.py Normal file
View File

@@ -0,0 +1,12 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py',
'../_base_/datasets/cityscapes_769x769.py', '../_base_/default_runtime.py',
'../_base_/schedules/schedule_80k.py'
]
crop_size = (769, 769)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(align_corners=True),
auxiliary_head=dict(align_corners=True),
test_cfg=dict(mode='slide', crop_size=(769, 769), stride=(513, 513)))

10
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb4-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,10 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py', '../_base_/datasets/ade20k.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
]
crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(mask_size=(66, 66), num_classes=150),
auxiliary_head=dict(num_classes=150))

11
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb4-20k_voc12aug-512x512.py Normal file
View File

@@ -0,0 +1,11 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py',
'../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
'../_base_/schedules/schedule_20k.py'
]
crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(num_classes=21),
auxiliary_head=dict(num_classes=21))

11
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb4-40k_voc12aug-512x512.py Normal file
View File

@@ -0,0 +1,11 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py',
'../_base_/datasets/pascal_voc12_aug.py', '../_base_/default_runtime.py',
'../_base_/schedules/schedule_40k.py'
]
crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(num_classes=21),
auxiliary_head=dict(num_classes=21))

10
Seg_All_In_One_MMSeg/configs/psanet/psanet_r50-d8_4xb4-80k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,10 @@
_base_ = [
'../_base_/models/psanet_r50-d8.py', '../_base_/datasets/ade20k.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_80k.py'
]
crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)
model = dict(
data_preprocessor=data_preprocessor,
decode_head=dict(mask_size=(66, 66), num_classes=150),
auxiliary_head=dict(num_classes=150))