Single-image super-resolution refers to the problem of generating a high-resolution image from a low-resolution one. In this work we address to the problem of single-image super-resolution of degraded low-resolution images. We face this task by adopting Convolutional Neural Networks, in particular we combine unsupervised adversarial learning with techniques aimed to preserve the color information and the spatial smoothness on the produced images.

Image Super-Resolution with Adversarial Learning

Boem, Davide
2019/2020

Abstract

Single-image super-resolution refers to the problem of generating a high-resolution image from a low-resolution one. In this work we address to the problem of single-image super-resolution of degraded low-resolution images. We face this task by adopting Convolutional Neural Networks, in particular we combine unsupervised adversarial learning with techniques aimed to preserve the color information and the spatial smoothness on the produced images.
2019-10-15
super-resolution, deep learning, networks, denoising, cyclegan
File in questo prodotto:
File Dimensione Formato  
davide_boem_tesi.pdf

accesso aperto

Dimensione 2.26 MB
Formato Adobe PDF
2.26 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/24606