S2ED: From Story to Executable Descriptions for Consistency-Aware Story Illustration
For story illustration, S2ED offers a training-free method to maintain long-horizon coherence, addressing a key bottleneck in multi-frame generation.
S2ED converts full stories into executable descriptions for consistent multi-frame illustration, improving sequence-level consistency and character fidelity over strong baselines on Flintstones and Shakoo Maku datasets.
Multi-frame story illustration requires long-horizon coherence beyond single-image text-to-image generation, including narrative decomposition and persistent character identity, layout, and affect across frames. We propose Story-to-Executable Descriptions (S2ED), a training-free, model-agnostic, prompt-layer framework that converts a full story into a sequence of explicit, editable executable descriptions for more consistent rendering. S2ED coordinates three agents to segment the narrative, ground canonical character attributes, and enrich spatial and affective cues, enabling interpretable prompt-carried state propagation and local edits to repair drift without retraining the generator. Experiments on Flintstones and Shakoo Maku show that S2ED improves sequence-level consistency and character fidelity over strong prompting, large-model planning, and a reference training-based method, under both automatic metrics and human judgments. We also deploy S2ED in an end-to-end story-to-storybook system for children's illustrated stories, with a supplementary video.