Latent Painter
This work addresses a creative art problem for generative AI users by providing a novel animation method, though it appears incremental as it builds on existing latent diffuser technology.
The authors tackled the limitation of latent diffusers in generating only sharpening animations by introducing Latent Painter, which uses the latent as a canvas and diffuser predictions as a plan to create painting animations and enable transitions between images from different checkpoints.
Latent diffusers revolutionized the generative AI and inspired creative art. When denoising the latent, the predicted original image at each step collectively animates the formation. However, the animation is limited by the denoising nature of the diffuser, and only renders a sharpening process. This work presents Latent Painter, which uses the latent as the canvas, and the diffuser predictions as the plan, to generate painting animation. Latent Painter also transits one generated image to another, which can happen between images from two different sets of checkpoints.