OAPT: Offset-Aware Partition Transformer for Double JPEG Artifacts Removal

Qiao Mo, Yukang Ding, Jinhua Hao, Qiang Zhu, Ming Sun, Chao Zhou, Feiyu Chen, Shuyuan Zhu

Official implementation of OAPT in ECCV2024, which is a transformer-based network designed for double (or multiple) compressed image restoration.

Architecture

Pattern clustering & inv operation

Experimental results on gray double JPEG images

Visual results

Training details

All the weights are put in Baidu Netdisk and Gdrive

Model(Gray)	Params(M)	Multi-Adds(G)	TrainingSets	Pretrain model	iterations
SwinIR	11.49	293.42	DF2K	006_CAR_DFWB_s126w7_SwinIR-M_jpeg10	200k
HAT-S	9.24	227.14	DF2K	HAT-S_SRx2	800k
ART	16.14	415.51	DF2K	CAR_ART_q10	200k
OAPT	12.96	293.60	DF2K	006_CAR_DFWB_s126w7_SwinIR-M_jpeg10	200k

Setup

The version of PyTorch we used is 1.7.0. Please ensure you have the correct versions of all dependencies by installing from the requirements.txt file.

pip install -r requirements.txt
python setup.py develop

Train

CUDA_VISIBLE_DEVICES=0,1,2,3 python -m torch.distributed.launch --nproc_per_node=4 --master_port=73 oapt/train.py -opt options/Gray/train/train_oapt.yml --launcher pytorch

Test

CUDA_VISIBLE_DEVICES=0 python oapt/test.py -opt ./options/Gray/test/test_oapt.yml

Acknowledgement

This project is mainly based on SwinIR and HAT.