KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #ai-agent #developer-tools #codex #typescript #cli #deepseek-harness #claude-code #ai-agents #mcp #rust

EDAM

★ 39 Open GitHub ↗

The code and pretrain models of EDAM-WSSS(Embedded Discriminative Attention Mechanism for Weakly Supervised Semantic Segmentation)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Embedded Discriminative Attention Mechanism for Weakly Supervised Semantic Segmentation(EDAM-WSSS) CVPR2021
The code and pretrain models of EDAM(Embedded Discriminative Attention Mechanism for Weakly Supervised Semantic Segmentation)
Our papaer [link](https://openaccess.thecvf.com/content/CVPR2021/papers/Wu_Embedded_Discriminative_Attention_Mechanism_for_Weakly_Supervised_Semantic_Segmentation_CVPR_2021_paper.pdf)
## Env
We train our model with Python 3.5, PyTorch 1.1.0 and 4 Tesla V100 GPUs with 16 GB memory. Other Python modules can be installed by running
 ```
conda install --yes --file requirements.txt
 ```
You can also download our env directly [[BaiduYun]](https://pan.baidu.com/s/18oMN2_1gAbmdFTaJBDEL5A) tlu7

After unzipping, running
 ```
source py3.5/bin/activate
 ```
 
## Dataset
  * [The PASCAL VOC 2012 development kit](http://host.robots.ox.ac.uk/pascal/VOC/voc2012/): You also need to specify the path ('voc12_root') of your downloaded dev kit.
  * Saliency Maps [[GoogleDrive]](https://drive.google.com/file/d/1ENS6jR6EUIDtxWsYwgwZ9YELSp0tFQ0k/view?usp=sharing)[[BaiduYun]](https://pan.baidu.com/s/1LfTwcm22Zup84yB635Ij1g) (fjst): Class-agnostic saliency maps generated by [PoolNet](https://github.com/backseason/PoolNet) as background cues.

## Pretrain Models
We provide three pre-trained models, including initialization parameters, EDAM parameters and Deeplabv2 parameters [[Google Drive]](https://drive.google.com/file/d/1Ng58UOStjRZ2DubePY_DYo-B2VT_LHid/view?usp=sharing)[[BaiduYun]](https://pan.baidu.com/s/1hbhBQKFAWtE69mucJXf6wQ) u7q7

## Train
Training EDAM from scratch.(Noting: We have pre-processed the dataset, and the amount of data in each epoch is equivalent to the original 25 epochs)
```
python3 train_EDAM_cls.py --lr 0.001 --batch_size 4 --max_epoches 1 --crop_size 368 --network network.resnet38_EDAM_cls --voc12_root [Root of VOC12] --weights [Root of initialization parameters] --wt_dec 5e-4 --session_name resnet38_EDAM_cls
```

To monitor loss and lr, run the following command in a separate terminal.
```
tensorboard --logdir runs
```

## Test
Generate pseudo labels.
 ```
python3 infer_cls.py --infer_list voc12/train_aug.txt --voc12_root /workdir/VOCdevkit/VOC2012 --network network.resnet38_EDAM_cls --weights [Path of EDAM Parameters]  --out_crf_pred [Output Path] --theta 0.2  --alpha 0.000001  --beta 0.99999 --sal_path [Path of Saliency Map]
 ```
 
Vis pseudo labels 
 ```
python3 colorful.py --img_path [Path of Pseudo Labels] —out_path [Path of Colorful Pseudo Labels]
 ```
 
## Segmentation Network
We use our pseudo labels fully-supervised train a [Deeplab-v2 Network](https://github.com/kazuto1011/deeplab-pytorch)  
We also provide the final pseudo labels for segmentation network training. [[Google Drive]](https://drive.google.com/file/d/1UkCIYI8Ink2S8l7aOue0kKQJMgkoCbOg/view?usp=sharing)[[BaiduYun]](https://pan.baidu.com/s/1ovEYet0JTiW9wj8UuP7-0g) 9aij

## Results
<table>
 <tr>
 <th>Model</th>
 <th>Train set</th>
 <th>Val set</th>
 <th>Crf?</th>
 <th>Saliency?</th>
 <th>Mean IoU</th>
 </tr>
 <tr>
 <td rowspan="3">EDAM</td>
 <td rowspan="3">
 trainaug
 </td>
 <td rowspan="3">
 train
 </td>
 <td>-</td>
 <td>-</td>
 <td>52.83</td>
 </tr>
 <tr>
 <td>&#10003;</td>
 <td>-</td>
 <td>58.61</td>
 </tr>
 <tr>
 <td>&#10003;</td>
 <td>&#10003;</td>
 <td>68.11</td>
 </tr>
 <tr>
 <td rowspan="3">DeepLab-v2</td>
 <td rowspan="3">
 trainaug
 </td>
 <td rowspan="3">
 val
 </td>
 <td>-</td>
 <td>-</td>
 <td>69.66</td>
 </tr>
 <tr>
 <td>&#10003;</td>
 <td>-</td>
 <td>70.96</td>
 </tr>
</table>

## Citation
If this code is useful to you, consider using the following citation:
```
@InProceedings{Wu_2021_CVPR,
    author    = {Wu, Tong and Huang, Junshi and Gao, Guangyu and Wei, Xiaoming and Wei, Xiaolin and Luo, Xuan and Liu, Chi Harold},
    title     = {Embedded Discriminative Attention Mechanism for Weakly Supervised Semantic Segmentation},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2021},
    pages     = {16765-16774}
}

EDAM

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README