Learning to Maintain Natural Image Statistics

By Roey Mechrez, Itamar Talmi, Firas Shama, Lihi Zelnik-Manor. 2018

33 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

The Perception-Distortion Tradeoff (2017, neutral): “[28].”
Image Inpainting via Generative Multi-column Convolutional Neural Networks (2018, methodology): “We instead adopt a relative distance measure [17, 16, 22] to model the relation between local features and target feature set.”
Analyzing Perception-Distortion Tradeoff using Enhanced Perceptual Super-resolution Network (2018, methodology): “Dataset Scores ENet[39] CX[34] To perform a region-wise comparison, we train both BNet and EPSR with a diﬀerent set of weights for MSE loss and adversarial loss.”
Deep Learning for Single Image Super-Resolution: A Brief Review (2018, methodology): “Based on nonparametric estimation, recently, the contextual loss [115], [116] was proposed by Mechrez et al.”
The 2018 PIRM Challenge on Perceptual Image Super-Resolution (2018, methodology): “These enabled to impressively surpass the performance of baselines, such as EnhanceNet [31] and CX [24].”
The Contextual Loss for Image Transformation with Non-Aligned Data (2018, methodology): “In our follow-up work, [34], we suggest to use the Contextual loss for realistic restoration, specifically for the tasks of super-resolution and surface normal estimation.”
The Unreasonable Effectiveness of Texture Transfer for Single Image Super-resolution (2018, methodology): “A new deep features based contextual loss [29] is used by [30] to maintain the natural image statistics of Iest.”
Creating High Resolution Images with a Latent Adversarial Generator (2020, background): “Still more losses can be added, for example a texture loss [15] or a contextual loss [13], leading to various perception-distortion trade-offs [2].”
2018 PIRM Challenge on Perceptual Image Super-resolution (2018, methodology): “These enabled to impressively surpass the performance of baselines, such as EnhanceNet [31] and CX [24].”