StoryTailor:A Zero-Shot Pipeline for Action-Rich Multi-Subject Visual Narratives

TL;DR

StoryTailor is a zero-shot pipeline that generates coherent, action-rich visual narratives from long prompts, maintaining subject identity and background continuity without fine-tuning, using innovative modules on a single high-end GPU.

Contribution

The paper introduces a novel zero-shot method with three modules for producing multi-subject visual narratives, addressing key challenges in action faithfulness and scene coherence.

Findings

CLIP-T improves by up to 15% over baselines

Inference is faster than FluxKontext on a 24 GB GPU

Qualitative results show expressive, stable scene generation

Abstract

Generating multi-frame, action-rich visual narratives without fine-tuning faces a threefold tension: action text faithfulness, subject identity fidelity, and cross-frame background continuity. We propose StoryTailor, a zero-shot pipeline that runs on a single RTX 4090 (24 GB) and produces temporally coherent, identity-preserving image sequences from a long narrative prompt, per-subject references, and grounding boxes. Three synergistic modules drive the system: Gaussian-Centered Attention (GCA) to dynamically focus on each subject core and ease grounding-box overlaps; Action-Boost Singular Value Reweighting (AB-SVR) to amplify action-related directions in the text embedding space; and Selective Forgetting Cache (SFC) that retains transferable background cues, forgets nonessential history, and selectively surfaces retained cues to build cross-scene semantic ties. Compared with baseline methods, experiments show that CLIP-T improves by up to 10-15%, with DreamSim lower than strong baselines, while CLIP-I stays in a visually acceptable, competitive range. With matched resolution and steps on a 24 GB GPU, inference is faster than FluxKontext. Qualitatively, StoryTailor delivers expressive interactions and evolving yet stable scenes.