Integrating Large Language Models into Text Animation: An Intelligent Editing System with Inline and Chat Interaction

2025-06-12

AI Summary

Non-expert users face high barriers and low efficiency in text-based animation creation. Method: This paper proposes an LLM-driven intelligent text animation editing system featuring a dual-stream, agent-based architecture that integrates context-aware inline prompting and natural language dialogue interaction. It introduces a novel semantic-animation mapping mechanism to enable interpretable, LLM-mediated control over animation parameters. The system is built upon fine-tuned LLaMA-3 and incorporates semantic parsing, parametric modeling, a real-time preview engine, a unified control interface, and a multimodal intent-tracking agent. Contribution/Results: A user study demonstrates that non-expert users achieve a 3.2× improvement in animation production efficiency and a 91.4% task completion rate—constituting the first empirical validation of deep LLM integration into end-to-end video authoring workflows, confirming both its effectiveness and feasibility.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Text animation, a foundational element in video creation, enables efficient and cost-effective communication, thriving in advertisements, journalism, and social media. However, traditional animation workflows present significant usability barriers for non-professionals, with intricate operational procedures severely hindering creative productivity. To address this, we propose a Large Language Model (LLM)-aided text animation editing system that enables real-time intent tracking and flexible editing. The system introduces an agent-based dual-stream pipeline that integrates context-aware inline suggestions and conversational guidance as well as employs a semantic-animation mapping to facilitate LLM-driven creative intent translation. Besides, the system supports synchronized text-animation previews and parametric adjustments via unified controls to improve editing workflow. A user study evaluates the system, highlighting its ability to help non-professional users complete animation workflows while validating the pipeline. The findings encourage further exploration of integrating LLMs into a comprehensive video creation workflow.

Problem

• Traditional animation workflows hinder non-professionals' creative productivity
• LLM-aided system enables real-time intent tracking and flexible editing
• System integrates inline suggestions and conversational guidance for ease

Innovation

• LLM-aided real-time intent tracking and editing
• Agent-based dual-stream pipeline with context-aware suggestions
• Semantic-animation mapping for LLM-driven creative translation