Learning Structral coherence Via Generative Adversarial Network for Single Image Super-Resolution

Abstract

Among the major remaining challenges for single image super resolution (SISR) is the capacity to recover coherent images with global shapes and local details conforming to human vision system. Recent generative adversarial network (GAN) based SISR methods have yielded overall realistic SR images, however, there are always unpleasant textures accompanied with structural distortions in local regions. To target these issues, we introduce the gradient branch into the generator to preserve structural information by restoring high-resolution gradient maps in SR process. In addition, we utilize a U-net based discriminator to consider both the whole image and the detailed per-pixel authenticity, which could encourage the generator to maintain overall coherence of the reconstructed images. Moreover, we have studied objective functions and LPIPS perceptual loss is added to generate more realistic and natural details. Experimental results show that our proposed method outperforms state-of-the-art perceptual-driven SR methods in perception index (PI), and obtains more geometrically consistent and visually pleasing textures in natural image restoration.

arXiv comment5 pages, 3 figures, 2 tables

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
• The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
• The relativistic discriminator: a key element missing from standard GAN
• Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform
• Meta-SR: A Magnification-Arbitrary Network for Super-Resolution
• Structure-Preserving Super Resolution with Gradient Guidance