Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models

arXiv:2311.17919v2 Announce Type: replace
Abstract: We address the problem of synthesizing multi-view optical illusions: images that change appearance upon a transformation, such as a flip or rotation. We propose a simple, zero-shot method for obtaining these illusions from off-the-shelf text-to-image diffusion models. During the reverse diffusion process, we estimate the noise from different views of a noisy image, and then combine these noise estimates together and denoise the image. A theoretical analysis suggests that this method works precisely for views …

arxiv cs.cv diffusion diffusion models optical type view visual