Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Xingjian, Chai, Li, Chen, Jiming
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2405.10014
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918148357226496
author Wang, Xingjian
Chai, Li
Chen, Jiming
author_facet Wang, Xingjian
Chai, Li
Chen, Jiming
contents The performance of single image super-resolution depends heavily on how to generate and complement high-frequency details to low-resolution images. Recently, diffusion-based DDPM models exhibit great potential in generating high-quality details for super-resolution tasks. They tend to directly predict high-frequency information of wide bandwidth by solely utilizing the high-resolution ground truth as the target for all sampling timesteps. However, as a result, they encounter hallucination problem that they generate mismatching artifacts. To tackle this problem and achieve higher-quality super-resolution, we propose a novel Frequency Domain-guided multiscale Diffusion model (FDDiff), which decomposes the high-frequency information complementing process into finer-grained steps. In particular, a wavelet packet-based frequency degradation pyramid is developed to provide multiscale intermediate targets with increasing bandwidth. Based on these targets, FDDiff guides reverse diffusion process to progressively complement missing high-frequency details over timesteps. Moreover, a multiscale frequency refinement network is designed to predict the required high-frequency components at multiple scales within one unified network. Comprehensive evaluations on popular benchmarks are conducted, and demonstrate that FDDiff outperforms prior generative methods with higher-fidelity super-resolution results.
format Preprint
id arxiv_https___arxiv_org_abs_2405_10014
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Frequency-Domain Refinement with Multiscale Diffusion for Super Resolution
Wang, Xingjian
Chai, Li
Chen, Jiming
Computer Vision and Pattern Recognition
Image and Video Processing
The performance of single image super-resolution depends heavily on how to generate and complement high-frequency details to low-resolution images. Recently, diffusion-based DDPM models exhibit great potential in generating high-quality details for super-resolution tasks. They tend to directly predict high-frequency information of wide bandwidth by solely utilizing the high-resolution ground truth as the target for all sampling timesteps. However, as a result, they encounter hallucination problem that they generate mismatching artifacts. To tackle this problem and achieve higher-quality super-resolution, we propose a novel Frequency Domain-guided multiscale Diffusion model (FDDiff), which decomposes the high-frequency information complementing process into finer-grained steps. In particular, a wavelet packet-based frequency degradation pyramid is developed to provide multiscale intermediate targets with increasing bandwidth. Based on these targets, FDDiff guides reverse diffusion process to progressively complement missing high-frequency details over timesteps. Moreover, a multiscale frequency refinement network is designed to predict the required high-frequency components at multiple scales within one unified network. Comprehensive evaluations on popular benchmarks are conducted, and demonstrate that FDDiff outperforms prior generative methods with higher-fidelity super-resolution results.
title Frequency-Domain Refinement with Multiscale Diffusion for Super Resolution
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2405.10014