Dynamic Typography: Bringing Text to Life via Video Diffusion Prior

TL;DR

This paper introduces 'Dynamic Typography,' an automated framework that animates text by deforming letters and adding motion based on user prompts, combining vector graphics and neural optimization to produce semantically meaningful and legible text animations.

Contribution

It presents a novel end-to-end method that integrates shape deformation and motion infusion for text animation, improving coherence and readability over baseline approaches.

Findings

Effective in generating coherent, semantically meaningful text animations

Outperforms baseline methods in quality and consistency

Demonstrates generalizability across various text-to-video models

Abstract

Text animation serves as an expressive medium, transforming static communication into dynamic experiences by infusing words with motion to evoke emotions, emphasize meanings, and construct compelling narratives. Crafting animations that are semantically aware poses significant challenges, demanding expertise in graphic design and animation. We present an automated text animation scheme, termed "Dynamic Typography", which combines two challenging tasks. It deforms letters to convey semantic meaning and infuses them with vibrant movements based on user prompts. Our technique harnesses vector graphics representations and an end-to-end optimization-based framework. This framework employs neural displacement fields to convert letters into base shapes and applies per-frame motion, encouraging coherence with the intended textual concept. Shape preservation techniques and perceptual loss regularization are employed to maintain legibility and structural integrity throughout the animation process. We demonstrate the generalizability of our approach across various text-to-video models and highlight the superiority of our end-to-end methodology over baseline methods, which might comprise separate tasks. Through quantitative and qualitative evaluations, we demonstrate the effectiveness of our framework in generating coherent text animations that faithfully interpret user prompts while maintaining readability. Our code is available at: https://animate-your-word.github.io/demo/.