Optical Instruments, Volume. 46, Issue 5, 1(2024)
Architectural style classification algorithm fusing CNN and Transformer
Figures & Tables(8)
Fig. 2. Structure of architectural style classification network（FCT-Net）
Fig. 4. Confusion matrix of partial results on Architectural Style Dataset
Table 1. Comparison of accuracy of different models on Architectural Style Dataset
View tableView in ArticleTable 1. Comparison of accuracy of different models on Architectural Style Dataset
模型 准确率/% 40%类别 100%类别 注：黑体为同类别中最大准确率 DCNN[ 6 ]72.42 66.60 MonuNet[ 24 ]71.20 61.93 ResNet-50 80.19 67.41 Inception-v3 67.15 60.06 ViT 70.01 57.14 Swin-Transformer 75.36 65.28 Visformer 76.33 70.49 FCT-Net（ours） 83.09 79.83
Table 2. Comparison of accuracy of different models on WikiChurches
View tableView in ArticleTable 2. Comparison of accuracy of different models on WikiChurches
模型 准确率/% 注：黑体为最大准确率。 MobileNet-V2 56.63 Swin-Transformer 52.05 Mobile-former 60.39 Conformer 63.50 Visformer 61.54 FCT-Net（ours） 68.41
Table 3. Comparison of accuracy of different types of models on public datasets
View tableView in ArticleTable 3. Comparison of accuracy of different types of models on public datasets
模型 准确率/% Architectural Style Dataset WikiChurches 注：黑体为最大准确率。 ResNet-50 67.41 62.36 MobileNet-V2 66.67 56.63 ViT 57.14 49.10 Swin-Transformer 65.28 52.05 FCT-Net（ours） 79.83 68.41
Table 4. Comparison of accuracy of CT-Block modules on the Architectural Style Dataset
View tableView in ArticleTable 4. Comparison of accuracy of CT-Block modules on the Architectural Style Dataset
模型 准确率/% Net1 75.91 Net2 63.69 FCT-Net 79.83
Tools
Get Citation
Copy Citation Text
Dong LIU, Rongfu ZHANG, Junxiang QIN, Junzhe GONG, Zhibin CAO. Architectural style classification algorithm fusing CNN and Transformer[J]. Optical Instruments, 2024, 46(5): 1
Paper Information
Category:
Received: Aug. 16, 2023
Accepted: --
Published Online: Jan. 3, 2025
The Author Email: ZHANG Rongfu (zrf@usst.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20