CABM: Content-Aware Bit Mapping for Single Image Super-Resolution Network with Large Input (CVPR 2023)

This repository is the official implementation of our CVPR2023 paper. paper.

Our implementation is based on CADyQ(PyTorch) and PAMS(PyTorch).

Due to the numerous settings in our paper, we have only provided a simplified version of training and testing code here.

Dependencies

• kornia (pip install kornia)
• Python >= 3.6
• PyTorch >= 1.10.0
• other packages used in our code

Datasets

• For training, we use DIV2K datasets.

• For testing, we use benchmark datasets and Test2K,4K.8K.

  # for training
  DIV2K 

  # for testing
  benchmark
  Test2K
  Test4K

How to train CABM step by step

# Taking EDSR as an example

# Step-1
# Train full-precision models
sh train_edsrbaseline_org.sh

# Step-2
# Train 8-bit PAMS models
sh train_edsrbaseline_pams.sh

# Step-3
# Train CADyQ models
sh train_edsrbaseline_cadyq.sh

# Step-4
# Get edge-to-bit tables
sh test_edsrbaseline_get_cabm_config.sh

# Step-5
# Get CABM models
sh train_edsrbaseline_cabm_simple.sh

How to test CABM

test_edsrbaseline_cabm_simple.sh

How to sample patches while training

You may refer to SamplingAUG.

Citation

@article{Tian2023CABMCB,
  title={CABM: Content-Aware Bit Mapping for Single Image Super-Resolution Network with Large Input},
  author={Senmao Tian and Ming Lu and Jiaming Liu and Yandong Guo and Yurong Chen and Shunli Zhang},
  journal={ArXiv},
  year={2023},
  volume={abs/2304.06454}
}