Prompting Gemini with damage-level definitions yields synthetic disaster imagery that a pre-trained classifier labels at F1 around 0.64, close to its score on real images, but the comparison lacks statistical support.
Introduction to the DDDAS2022 Conference Infosymbiotics/Dynamic Data Driven Applications Systems,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Effective Damage Data Generation by Fusing Imagery with Human Knowledge Using Vision-Language Models
Prompting Gemini with damage-level definitions yields synthetic disaster imagery that a pre-trained classifier labels at F1 around 0.64, close to its score on real images, but the comparison lacks statistical support.