Learning Arbitrary-Scale RAW Image Downscaling with Wavelet-based Recurrent Reconstruction

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Ren, Yang, Jiang, Hai, Li, Wei, Yang, Menglong, Zhang, Heng, Sheng, Zehua, Ye, Qingsheng, Liu, Shuaicheng
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911084857786368
author Ren, Yang
Jiang, Hai
Li, Wei
Yang, Menglong
Zhang, Heng
Sheng, Zehua
Ye, Qingsheng
Liu, Shuaicheng
author_facet Ren, Yang
Jiang, Hai
Li, Wei
Yang, Menglong
Zhang, Heng
Sheng, Zehua
Ye, Qingsheng
Liu, Shuaicheng
contents Image downscaling is critical for efficient storage and transmission of high-resolution (HR) images. Existing learning-based methods focus on performing downscaling within the sRGB domain, which typically suffers from blurred details and unexpected artifacts. RAW images, with their unprocessed photonic information, offer greater flexibility but lack specialized downscaling frameworks. In this paper, we propose a wavelet-based recurrent reconstruction framework that leverages the information lossless attribute of wavelet transformation to fulfill the arbitrary-scale RAW image downscaling in a coarse-to-fine manner, in which the Low-Frequency Arbitrary-Scale Downscaling Module (LASDM) and the High-Frequency Prediction Module (HFPM) are proposed to preserve structural and textural integrity of the reconstructed low-resolution (LR) RAW images, alongside an energy-maximization loss to align high-frequency energy between HR and LR domain. Furthermore, we introduce the Realistic Non-Integer RAW Downscaling (Real-NIRD) dataset, featuring a non-integer downscaling factor of 1.3$\times$, and incorporate it with publicly available datasets with integer factors (2$\times$, 3$\times$, 4$\times$) for comprehensive benchmarking arbitrary-scale image downscaling purposes. Extensive experiments demonstrate that our method outperforms existing state-of-the-art competitors both quantitatively and visually. The code and dataset will be released at https://github.com/RenYangSCU/ASRD.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23219
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Arbitrary-Scale RAW Image Downscaling with Wavelet-based Recurrent Reconstruction
Ren, Yang
Jiang, Hai
Li, Wei
Yang, Menglong
Zhang, Heng
Sheng, Zehua
Ye, Qingsheng
Liu, Shuaicheng
Image and Video Processing
Computer Vision and Pattern Recognition
Image downscaling is critical for efficient storage and transmission of high-resolution (HR) images. Existing learning-based methods focus on performing downscaling within the sRGB domain, which typically suffers from blurred details and unexpected artifacts. RAW images, with their unprocessed photonic information, offer greater flexibility but lack specialized downscaling frameworks. In this paper, we propose a wavelet-based recurrent reconstruction framework that leverages the information lossless attribute of wavelet transformation to fulfill the arbitrary-scale RAW image downscaling in a coarse-to-fine manner, in which the Low-Frequency Arbitrary-Scale Downscaling Module (LASDM) and the High-Frequency Prediction Module (HFPM) are proposed to preserve structural and textural integrity of the reconstructed low-resolution (LR) RAW images, alongside an energy-maximization loss to align high-frequency energy between HR and LR domain. Furthermore, we introduce the Realistic Non-Integer RAW Downscaling (Real-NIRD) dataset, featuring a non-integer downscaling factor of 1.3$\times$, and incorporate it with publicly available datasets with integer factors (2$\times$, 3$\times$, 4$\times$) for comprehensive benchmarking arbitrary-scale image downscaling purposes. Extensive experiments demonstrate that our method outperforms existing state-of-the-art competitors both quantitatively and visually. The code and dataset will be released at https://github.com/RenYangSCU/ASRD.
title Learning Arbitrary-Scale RAW Image Downscaling with Wavelet-based Recurrent Reconstruction
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2507.23219