Pixel Recursive Super Resolution
Ryan Dahl Mohammad Norouzi Jonathon Shlens Google Brain Work done as a member of the Google Brain Residency program (g.co/brainresidency).
Abstract
Super resolution is the problem of artificially enlarging a low resolution photograph to recover a plausible high resolution version. In the regime of high magnification factors, the problem is dramatically underspecified and many plausible, high resolution images may match a given low resolution image. In particular, traditional super resolution techniques fail in this regime due to the multimodality of the problem and strong prior information that must be imposed on image synthesis to produce plausible high resolution images. In this work we propose a new probabilistic deep network architecture, a pixel recursive super resolution model, that is an extension of PixelCNNs to address this problem. We demonstrate that this model produces a diversity of plausible high resolution images at large magnification factors. Furthermore, in human evaluation studies we demonstrate how previous methods fail to fool human observers. However, high resolution images sampled from this probabilistic deep network do fool a naive human observer a significant fraction of the time.
中文速览
把一张极低分辨率的照片（比如8×8像素的人脸）放大8倍恢复成高清图像，本质上是一个"一对多"的病态问题，传统方法因为只优化均方误差等单峰目标，输出要么模糊、要么无法反映多种合理可能性。研究者提出了"像素递归超分辨率模型"（Pixel Recursive Super Resolution），将条件生成网络ResNet与自回归模型PixelCNN结合起来，让模型在生成每个像素时同时参考低分辨率输入和已生成的周边像素，从而捕捉输出像素之间的复杂依赖关系，并以对数似然为目标端到端训练。实验结果显示，从该模型中采样可以生成多样且清晰的高分辨率图像，在人类评估研究中，这些样本有高达27.9%的概率被误判为真实照片，而现有最先进的深度学习方法几乎无法骗过人眼。这项工作表明，全概率自回归框架是解决高倍率超分辨率这类高度欠定问题的有效路径，也为条件图像生成模型的设计提供了新思路。
原文 arXiv:1702.00783;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1702.00783v2