BriefGPT.xyz
Aug, 2019
噪声流:带条件的标准化流噪声建模
Noise Flow: Noise Modeling with Conditional Normalizing Flows
HTML
PDF
Abdelrahman Abdelhamed, Marcus A. Brubaker, Michael S. Brown
TL;DR
介绍了基于normalizing flow的强大而精确的噪声模型Noise Flow,结合了基本参数噪声模型和神经网络的灵活性和表现力。与已有的噪声模型相比,Noise Flow具有显著的改进,是第一次试图超越简单的参数模型,利用深度学习和数据驱动的噪声分布来建模和综合图像噪声。
Abstract
Modeling and synthesizing
image noise
is an important aspect in many
computer vision
applications. The long-standing additive white Gaussian and heteroscedastic (signal-dependent) noise models widely used in the
→