Learning to Sample the Most Useful Training Patches from Images

by   Shuyang Sun, et al.

Some image restoration tasks like demosaicing require difficult training samples to learn effective models. Existing methods attempt to address this data training problem by manually collecting a new training dataset that contains adequate hard samples, however, there are still hard and simple areas even within one single image. In this paper, we present a data-driven approach called PatchNet that learns to select the most useful patches from an image to construct a new training set instead of manual or random selection. We show that our simple idea automatically selects informative samples out from a large-scale dataset, leading to a surprising 2.35dB generalisation gain in terms of PSNR. In addition to its remarkable effectiveness, PatchNet is also resource-friendly as it is applied only during training and therefore does not require any additional computational cost during inference.


page 1

page 2

page 7

page 8


Which Pixel to Annotate: a Label-Efficient Nuclei Segmentation Framework

Recently deep neural networks, which require a large amount of annotated...

Internal Diverse Image Completion

Image completion is widely used in photo restoration and editing applica...

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

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

IStego100K: Large-scale Image Steganalysis Dataset

In order to promote the rapid development of image steganalysis technolo...

A Multi-Armed Bandit to Smartly Select a Training Set from Big Medical Data

With the availability of big medical image data, the selection of an ade...

Blind Motion Deblurring through SinGAN Architecture

Blind motion deblurring involves reconstructing a sharp image from an ob...

Hashed Binary Search Sampling for Convolutional Network Training with Large Overhead Image Patches

Very large overhead imagery associated with ground truth maps has the po...

Please sign up or login with your details

Forgot password? Click here to reset