S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction

18 Oct 2018  ·  Bolun Zheng, Rui Sun, Xiang Tian, Yaowu Chen ·

Recent studies have used deep residual convolutional neural networks (CNNs) for JPEG compression artifact reduction. This study proposes a scalable CNN called S-Net. Our approach effectively adjusts the network scale dynamically in a multitask system for real-time operation with little performance loss. It offers a simple and direct technique to evaluate the performance gains obtained with increasing network depth, and it is helpful for removing redundant network layers to maximize the network efficiency. We implement our architecture using the Keras framework with the TensorFlow backend on an NVIDIA K80 GPU server. We train our models on the DIV2K dataset and evaluate their performance on public benchmark datasets. To validate the generality and universality of the proposed method, we created and utilized a new dataset, called WIN143, for over-processed images evaluation. Experimental results indicate that our proposed approach outperforms other CNN-based methods and achieves state-of-the-art performance.

PDF Abstract


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
JPEG Artifact Correction LIVE1 (Quality 10 Color) S-Net PSNR 27.35 # 5
PSNR-B 27.36 # 5
SSIM 0.809 # 4
JPEG Artifact Correction Live1 (Quality 10 Grayscale) S-Net PSNR 29.44 # 7
PSNR-B 29.39 # 3
SSIM 0.8325 # 5
JPEG Artifact Correction LIVE1 (Quality 20 Color) S-Net PSNR 29.81 # 4
PSNR-B 29.79 # 2
SSIM 0.878 # 3
JPEG Artifact Correction LIVE1 (Quality 20 Grayscale) S-Net PSNR 31.83 # 6
PSNR-B 31.76 # 3
SSIM 0.8975 # 4


No methods listed for this paper. Add relevant methods here