arXiv cs.CL· Yanir Harel, Shalom Rosner, Ronit D. Gross, Tal Halevi, Shilo Glam, Ella Koresh, Ido Kanter·· 11 小时前AI 评分20
频域分类为何优于空间域：VGG 与 ViT 混合模型研究
When frequency domain classification is superior to spatial
AI 导读
研究发现频域图像分类精度下降源于缺乏局部特征，在频域中局部交换数据会显著降低其逆空间域精度，但频域精度基本不变。使用空间域含全局特征的数据集可提升频域精度。据此构建的 VGG 与 ViT 混合模型（分别仅在频域和空间域训练）在 ImageNet-100、WikiArt、DTD、FMD、PSVRT 和 KTH-TIPS2-b 数据集上取得更高分类精度。
正文 · 原文
Abstract:Modern image classification models largely rely on local characteristics and feature extraction for their predictions. This focused approach works well in the spatial domain, but reportedly falls short in the frequency domain, even though both representations contain the same information. This study demonstrates that the significant reduction in accuracy in the frequency domain stems from the lack of local features. Locally swapping data in the frequency domain significantly decreases its inverse spatial domain accuracy, even below that of the frequency domain; however, the frequency domain accuracy remains relatively constant. Moreover, the use of datasets containing global features in the spatial domain yields higher accuracy in the frequency domain. Using these results, a hybrid model is constructed, combining one model trained only in the frequency domain and another trained only in the spatial domain to achieve higher classification accuracy. The proposed visual geometry group (VGG) and vision transformer (ViT) architectures are trained on the ImageNet-100, WikiArt, DTD, FMD, PSVRT, and KTH-TIPS2-b datasets and evaluated.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.00073 [cs.CV] |
| (or arXiv:2611.00073v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.00073 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shalom Rosner [view email]
[v1]
Thu, 15 Oct 2026 18:39:55 UTC (795 KB)
来源：arXiv cs.CL · arxiv.org