The Art of Deception: Color Visual Illusions and Diffusion Models
This work addresses the fundamental question of why both human brains and artificial neural networks are susceptible to the same visual illusions, with implications for understanding visual information processing in AI.
The study investigated whether diffusion models perceive visual illusions like humans, finding that they exhibit human-like brightness/color shifts in their latent space and can predict and generate new illusions that fool humans in psychophysical experiments.
Visual illusions in humans arise when interpreting out-of-distribution stimuli: if the observer is adapted to certain statistics, perception of outliers deviates from reality. Recent studies have shown that artificial neural networks (ANNs) can also be deceived by visual illusions. This revelation raises profound questions about the nature of visual information. Why are two independent systems, both human brains and ANNs, susceptible to the same illusions? Should any ANN be capable of perceiving visual illusions? Are these perceptions a feature or a flaw? In this work, we study how visual illusions are encoded in diffusion models. Remarkably, we show that they present human-like brightness/color shifts in their latent space. We use this fact to demonstrate that diffusion models can predict visual illusions. Furthermore, we also show how to generate new unseen visual illusions in realistic images using text-to-image diffusion models. We validate this ability through psychophysical experiments that show how our model-generated illusions also fool humans.