.png){kind=icon}
.png){kind=icon}
Return
Domain-adaptive generative data augmentation for building façade classification: Linking synthetic data to performance across urban contexts
.png){kind=icon}
OA
AI
Y
S
N
S
DOI：10.1016/j.dibe.2026.100991.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes a domain-adaptive generative augmentation framework for façade material classification under limited data conditions. • Enhances the realism and diversity of synthetic façade data by incorporating architectural and material-specific characteristics. • Quantifies the impact of synthetic data quality on downstream classification performance. • Improves classification accuracy, increasing from 85.00% to 93.67% (Korea) and from 71.67% to 83.67% (UK). • Demonstrates the practical applicability of the proposed domain-adaptive generative augmentation framework.
Keywords:
Generative augmentation
Diffusion models
Low-rank adaptation
Synthetic data
Facade materials
Deep learning
Data classification
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
8.2
Papers:
985
Citations:
3.3K
Organization
Cited Papers