Multiresolution Neural Networks for Imaging

We present MR-Net, a general architecture for multiresolution neural networks, and a framework for imaging applications based on this architecture. Our coordinate-based networks are continuous both in space and in scale as they are composed of multiple stages that progressively add finer details. Be...

Full description

Saved in:
Bibliographic Details
Main Authors: Paz, Hallison, Novello, Tiago, Silva, Vinicius, Schardong, Guilherme, Schirmer, Luiz, Chagas, Fabio, Lopes, Helio, Velho, Luiz
Format: Conference Proceeding
Language:English
Subjects:
Graphics
Image representation
Image resolution
Imaging
Level fo Detail
Multiresolution
Neural networks
Textures
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present MR-Net, a general architecture for multiresolution neural networks, and a framework for imaging applications based on this architecture. Our coordinate-based networks are continuous both in space and in scale as they are composed of multiple stages that progressively add finer details. Besides that, they are a compact and efficient representation. We show examples of multiresolution image representation and applications to texture magnification, minification, and antialiasing.
ISSN:2377-5416
DOI:10.1109/SIBGRAPI55357.2022.9991765