Scalability in Building Component Data Annotation: Enhancing Facade Material Classification with Synthetic Data
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929312444186624 |
|---|---|
| author | Harrison, Josie Hollberg, Alexander Yu, Yinan |
| author_facet | Harrison, Josie Hollberg, Alexander Yu, Yinan |
| contents | Computer vision models trained on Google Street View images can create material cadastres. However, current approaches need manually annotated datasets that are difficult to obtain and often have class imbalance. To address these challenges, this paper fine-tuned a Swin Transformer model on a synthetic dataset generated with DALL-E and compared the performance to a similar manually annotated dataset. Although manual annotation remains the gold standard, the synthetic dataset performance demonstrates a reasonable alternative. The findings will ease annotation needed to develop material cadastres, offering architects insights into opportunities for material reuse, thus contributing to the reduction of demolition waste. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_08557 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Scalability in Building Component Data Annotation: Enhancing Facade Material Classification with Synthetic Data Harrison, Josie Hollberg, Alexander Yu, Yinan Computer Vision and Pattern Recognition Machine Learning Computer vision models trained on Google Street View images can create material cadastres. However, current approaches need manually annotated datasets that are difficult to obtain and often have class imbalance. To address these challenges, this paper fine-tuned a Swin Transformer model on a synthetic dataset generated with DALL-E and compared the performance to a similar manually annotated dataset. Although manual annotation remains the gold standard, the synthetic dataset performance demonstrates a reasonable alternative. The findings will ease annotation needed to develop material cadastres, offering architects insights into opportunities for material reuse, thus contributing to the reduction of demolition waste. |
| title | Scalability in Building Component Data Annotation: Enhancing Facade Material Classification with Synthetic Data |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2404.08557 |