Animate Sketch: Text-Driven Sketch Animation Generation

Sketches represent a significant form of human visual expression. During the animation process, key frames and draft animations are indispensable components. However, this production process is time-consuming and demanding, requiring animator with substantial experience. In this paper, we propose an innovative algorithm that animates sketches based on the input text prompt. Concretely, the sketches are represented using Bezier curves, and the output is a vector sketch animation. Our model guides the placement of strokes with an optimized score distillation sampling (SDS) method based on a large pretrained text-to-video diffusion model. Moreover, the model incorporates geometric and semantic simplifications in its generation process. The results demonstrate that the sketch animations produced by our method yield favorable outcomes.