DeepAI AI Chat
Log In Sign Up

Deep Learning with Inaccurate Training Data for Image Restoration

by   Bolin Liu, et al.
Shanghai Jiao Tong University
McMaster University

In many applications of deep learning, particularly those in image restoration, it is either very difficult, prohibitively expensive, or outright impossible to obtain paired training data precisely as in the real world. In such cases, one is forced to use synthesized paired data to train the deep convolutional neural network (DCNN). However, due to the unavoidable generalization error in statistical learning, the synthetically trained DCNN often performs poorly on real world data. To overcome this problem, we propose a new general training method that can compensate for, to a large extent, the generalization errors of synthetically trained DCNNs.


page 1

page 5

page 6

page 7

page 8


Training Patch Analysis and Mining Skills for Image Restoration Deep Neural Networks

There have been numerous image restoration methods based on deep convolu...

Mediated Uncoupled Learning: Learning Functions without Direct Input-output Correspondences

Ordinary supervised learning is useful when we have paired training data...

Data–driven Image Restoration with Option–driven Learning for Big and Small Astronomical Image Datasets

Image restoration methods are commonly used to improve the quality of as...

The Deep Bootstrap: Good Online Learners are Good Offline Generalizers

We propose a new framework for reasoning about generalization in deep le...

Simulating analogue film damage to analyse and improve artefact restoration on high-resolution scans

Digital scans of analogue photographic film typically contain artefacts ...

Targeted Deep Learning: Framework, Methods, and Applications

Deep learning systems are typically designed to perform for a wide range...