Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2410.17255 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910661179604992 |
|---|---|
| author | Liu, Zilong Janowicz, Krzysztof Currier, Kitty Shi, Meilin |
| author_facet | Liu, Zilong Janowicz, Krzysztof Currier, Kitty Shi, Meilin |
| contents | Regional defaults describe the emerging phenomenon that text-to-image (T2I) foundation models used in generative AI are prone to over-proportionally depicting certain geographic regions to the exclusion of others. In this work, we introduce a scalable evaluation for uncovering such regional defaults. The evaluation consists of region hierarchy--based image generation and cross-level similarity comparisons. We carry out an experiment by prompting DALL-E 2, a state-of-the-art T2I generation model capable of generating photorealistic images, to depict a forest. We select forest as an object class that displays regional variation and can be characterized using spatial statistics. For a region in the hierarchy, our experiment reveals the regional defaults implicit in DALL-E 2, along with their scale-dependent nature and spatial relationships. In addition, we discover that the implicit defaults do not necessarily correspond to the most widely forested regions in reality. Our findings underscore a need for further investigation into the geography of T2I generation and other forms of generative AI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_17255 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Uncovering Regional Defaults from Photorealistic Forests in Text-to-Image Generation with DALL-E 2 Liu, Zilong Janowicz, Krzysztof Currier, Kitty Shi, Meilin Computers and Society Computer Vision and Pattern Recognition Machine Learning Regional defaults describe the emerging phenomenon that text-to-image (T2I) foundation models used in generative AI are prone to over-proportionally depicting certain geographic regions to the exclusion of others. In this work, we introduce a scalable evaluation for uncovering such regional defaults. The evaluation consists of region hierarchy--based image generation and cross-level similarity comparisons. We carry out an experiment by prompting DALL-E 2, a state-of-the-art T2I generation model capable of generating photorealistic images, to depict a forest. We select forest as an object class that displays regional variation and can be characterized using spatial statistics. For a region in the hierarchy, our experiment reveals the regional defaults implicit in DALL-E 2, along with their scale-dependent nature and spatial relationships. In addition, we discover that the implicit defaults do not necessarily correspond to the most widely forested regions in reality. Our findings underscore a need for further investigation into the geography of T2I generation and other forms of generative AI. |
| title | Uncovering Regional Defaults from Photorealistic Forests in Text-to-Image Generation with DALL-E 2 |
| topic | Computers and Society Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2410.17255 |