CubeFormer: A Simple yet Effective Baseline for Lightweight Image Super-Resolution

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Jikai, Zheng, Huan, Shen, Jianbing
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915045923880960
author Wang, Jikai
Zheng, Huan
Shen, Jianbing
author_facet Wang, Jikai
Zheng, Huan
Shen, Jianbing
contents Lightweight image super-resolution (SR) methods aim at increasing the resolution and restoring the details of an image using a lightweight neural network. However, current lightweight SR methods still suffer from inferior performance and unpleasant details. Our analysis reveals that these methods are hindered by constrained feature diversity, which adversely impacts feature representation and detail recovery. To respond this issue, we propose a simple yet effective baseline called CubeFormer, designed to enhance feature richness by completing holistic information aggregation. To be specific, we introduce cube attention, which expands 2D attention to 3D space, facilitating exhaustive information interactions, further encouraging comprehensive information extraction and promoting feature variety. In addition, we inject block and grid sampling strategies to construct intra-cube transformer blocks (Intra-CTB) and inter-cube transformer blocks (Inter-CTB), which perform local and global modeling, respectively. Extensive experiments show that our CubeFormer achieves state-of-the-art performance on commonly used SR benchmarks. Our source code and models will be publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2412_02234
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CubeFormer: A Simple yet Effective Baseline for Lightweight Image Super-Resolution
Wang, Jikai
Zheng, Huan
Shen, Jianbing
Computer Vision and Pattern Recognition
Lightweight image super-resolution (SR) methods aim at increasing the resolution and restoring the details of an image using a lightweight neural network. However, current lightweight SR methods still suffer from inferior performance and unpleasant details. Our analysis reveals that these methods are hindered by constrained feature diversity, which adversely impacts feature representation and detail recovery. To respond this issue, we propose a simple yet effective baseline called CubeFormer, designed to enhance feature richness by completing holistic information aggregation. To be specific, we introduce cube attention, which expands 2D attention to 3D space, facilitating exhaustive information interactions, further encouraging comprehensive information extraction and promoting feature variety. In addition, we inject block and grid sampling strategies to construct intra-cube transformer blocks (Intra-CTB) and inter-cube transformer blocks (Inter-CTB), which perform local and global modeling, respectively. Extensive experiments show that our CubeFormer achieves state-of-the-art performance on commonly used SR benchmarks. Our source code and models will be publicly available.
title CubeFormer: A Simple yet Effective Baseline for Lightweight Image Super-Resolution
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.02234