Efficient Cost-and-Quality Controllable Arbitrary-scale Super-resolution with Fourier Constraints

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Akita, Kazutoshi, Ukita, Norimichi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914117783126016
author Akita, Kazutoshi
Ukita, Norimichi
author_facet Akita, Kazutoshi
Ukita, Norimichi
contents Cost-and-Quality (CQ) controllability in arbitrary-scale super-resolution is crucial. Existing methods predict Fourier components one by one using a recurrent neural network. However, this approach leads to performance degradation and inefficiency due to independent prediction. This paper proposes predicting multiple components jointly to improve both quality and efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2510_23978
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Cost-and-Quality Controllable Arbitrary-scale Super-resolution with Fourier Constraints
Akita, Kazutoshi
Ukita, Norimichi
Computer Vision and Pattern Recognition
Cost-and-Quality (CQ) controllability in arbitrary-scale super-resolution is crucial. Existing methods predict Fourier components one by one using a recurrent neural network. However, this approach leads to performance degradation and inefficiency due to independent prediction. This paper proposes predicting multiple components jointly to improve both quality and efficiency.
title Efficient Cost-and-Quality Controllable Arbitrary-scale Super-resolution with Fourier Constraints
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.23978