• Team Name: ClarifyAI ( Team Id: Qual-230517)
• Rank 4

• Anupam Rawat
• Rahul Choudhary
• Shorya Sethia
• Suraj Prasad

Check out summary of our work in this report

• 🏠 Dependencies and installation
• 👏 How to run
  • 1️⃣ Download Pre-trained Models
  • 2️⃣Prepare Dataset
    • Training Dataset
    • Testing Dataset
    • Dataset Architectures
  • 3️⃣ Train
  • 4️⃣ Test
  • 5️⃣ Evaluate the Robustness of Our model on All dataset
• ☺️ Results

• Python >= 3.8.5
• PyTorch >= 1.11
• CUDA >= 11.3
• Other required packages in requirements.txt

# git clone this repository
git clone https://github.com/shoryasethia/RobustSIRR.git
cd RobustSIRR

# create new anaconda env
conda create -n sirr python=3.8 -y
conda activate sirr

# install python dependencies by pip
pip install -r requirements.txt

🌟 Download the pre-trained RobustSIRR models from [Pre-trained_RobustSIRR_BaiduYunDisk (pwd:sirr), Google Drvie] to the checkpoints folder.

• 7,643 cropped images with size 224 × 224 from Pascal VOC dataset (image ids are provided in VOC2012_224_train_png.txt, you should crop the center region with size 224 x 224 to reproduce our result )
• 90(89) real-world training images from Berkeley real dataset

❗ Place the processed VOC2012 and real datasets in the datasets folder, and name them VOC2012 and real89 respectively.

🌟For convenience, you can directly download the prepared training dataset from [ VOC2012_For_RobustSIRR_BaiduYunDisk (pwd:sirr) , Google Drvie] and [ real89_For_RobustSIRR_BaiduYunDisk (pwd:sirr) , Google Drvie].

• 20 real testing images from Berkeley real dataset
• Three sub-datasets, namely ‘Objects’, ‘Postcard’, ‘Wild’ from SIR² dataset
• 20 testing images from Nature

❗ Place the processed datasets in the datasets folder, and name them as real20, SIR2, and nature20 respectively.

🌟For convenience, you can directly download the prepared testing dataset from [ TestingDataset_For_RobustSIRR_BaiduYunDisk (pwd:sirr) , Google Drvie].

The hierarchical structure of all datasets is illustrated in the following diagram.

datasets
├── nature20
│   ├── blended
│   └── transmission_layer
├── real20
│   ├── blended
│   ├── real_test.txt
│   └── transmission_layer
├── real89
│   ├── blended
│   └── transmission_layer
├── SIR2
│   ├── PostcardDataset
│   │   ├── blended
│   │   ├── reflection
│   │   └── transmission_layer
│   ├── SolidObjectDataset
│   │   ├── blended
│   │   ├── reflection
│   │   └── transmission_layer
│   └── WildSceneDataset
│       ├── blended
│       ├── reflection
│       └── transmission_layer
└── VOC2012
    ├── blended
    ├── JPEGImages
    ├── reflection_layer
    ├── reflection_mask_layer
    ├── transmission_layer
    └── VOC_results_list.json

Note:

• transmission_layer is GT, blended is Input, and reflection/reflection_layer is the reflection part
• For the SIR^2 dataset, we only standardize the folder structure

• For adv. training:

# To Be Released

• For clean images training:

# ours_cvpr
CUDA_VISIBLE_DEVICES=0 python train.py --name ours --gpu_id 0 --no-verbose --display_id -1 --batchSize 4

# ours_wo_aid 
CUDA_VISIBLE_DEVICES=0 python train.py --name ours_wo_aid --gpu_id 0 --no-verbose --display_id -1 --batchSize 4 --wo_aid

# ours_wo_aff
CUDA_VISIBLE_DEVICES=0 python train.py --name ours_wo_aff --gpu_id 0 --no-verbose --display_id -1 --batchSize 4 --wo_aff

# ours_wo_scm
CUDA_VISIBLE_DEVICES=0 python train.py --name ours_wo_scm --gpu_id 0 --no-verbose --display_id -1 --batchSize 4 --wo_scm

Note:

• Check options/robustsirr/train_options.py to see more training options.

CUDA_VISIBLE_DEVICES=0 python test.py --name ours_cvpr --hyper --gpu_ids 0 -r --no-verbose --save_gt --save_attack --save_results

# To Be Released
# Due to confidentiality concerns. Alternatively, you can refer to https://github.com/yuyi-sd/Robust_Rain_Removal

☝️ Comparison of the PSNR values with respect to perturbation levels $\epsilon$ for different attacks on various datasets. ‘MSE FR Nature’ represents attacking on Full Region with MSE objective on the Nature dataset, and so the others.

☝️ Comparison of different training strategies on three benchmark datasets. ‘w/’ and ‘w/o adv.’ mean training with or without adversarial images. MSE and LPIPS denote corresponding attacks over Full regions. ↓ and ↑ represent the degradation and improvement performance compared to the original prediction inputting clean images.