A new 19,000-image photorealistic color illusion dataset shows that vision-language models are partially fooled by color illusions, producing human-aligned answers more often than chance but rarely matching human perception.
Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Evaluating Model Perception of Color Illusions in Photorealistic Scenes
A new 19,000-image photorealistic color illusion dataset shows that vision-language models are partially fooled by color illusions, producing human-aligned answers more often than chance but rarely matching human perception.