One-to-Many Network for Visually Pleasing Compression Artifacts Reduction

Abstract

We consider the compression artifacts reduction problem, where a compressed image is transformed into an artifact-free image. Recent approaches for this problem typically train a one-to-one mapping using a per-pixel $L_2$ loss between the outputs and the ground-truths. We point out that these approaches used to produce overly smooth results, and PSNR doesn't reflect their real performance. In this paper, we propose a one-to-many network, which measures output quality using a perceptual loss, a naturalness loss, and a JPEG loss. We also avoid grid-like artifacts during deconvolution using a "shift-and-average" strategy. Extensive experimental results demonstrate the dramatic visual improvement of our approach over the state of the arts.

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution
• Learning Deconvolution Network for Semantic Segmentation
• Identity Mappings in Deep Residual Networks
• On learning optimized reaction diffusion processes for effective image restoration

Cited by in corpus (4)

• Path-Restore: Learning Network Path Selection for Image Restoration
• Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning
• Attention-guided Image Compression by Deep Reconstruction of Compressive Sensed Saliency Skeleton
• Vehicle Image Generation Going Well with The Surroundings