LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping
This work addresses the challenge of creating interpretable anamorphic images for applications in art and visual effects, representing an incremental extension of prior work like Visual Anagrams.
The paper tackles the problem of generating anamorphic images that are distorted but still recognizable when viewed directly, using a method based on latent rectified flow models and Laplacian Pyramid Warping, resulting in high-quality visuals and novel perceptual illusions.
Anamorphosis refers to a category of images that are intentionally distorted, making them unrecognizable when viewed directly. Their true form only reveals itself when seen from a specific viewpoint, which can be through some catadioptric device like a mirror or a lens. While the construction of these mathematical devices can be traced back to as early as the 17th century, they are only interpretable when viewed from a specific vantage point and tend to lose meaning when seen normally. In this paper, we revisit these famous optical illusions with a generative twist. With the help of latent rectified flow models, we propose a method to create anamorphic images that still retain a valid interpretation when viewed directly. To this end, we introduce Laplacian Pyramid Warping, a frequency-aware image warping technique key to generating high-quality visuals. Our work extends Visual Anagrams (arXiv:2311.17919) to latent space models and to a wider range of spatial transforms, enabling the creation of novel generative perceptual illusions.