StoryTailor：面向动作丰富多主体视觉叙事的零样本流水线
StoryTailor:A Zero-Shot Pipeline for Action-Rich Multi-Subject Visual Narratives
GCA 用高斯加权注意力缓解多主体 grounding box 重叠，AB-SVR 在文本嵌入空间放大动作方向，SFC 缓存可迁移背景线索——三者都是可插拔的零样本模块，可直接迁移到多主体视频生成或长序列故事板生成中提升主体一致性与动作忠实度。
原文依据 Gaussian-Centered Attention (GCA) to dynamically focus on each subject core and ease grounding-box overlaps
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论在不进行微调的情况下生成多帧、动作丰富的视觉叙事面临三重张力：动作文本忠实度、主体身份保真度与跨帧背景连续性。我们提出 StoryTailor，一种零样本流水线，可在单块 RTX 4090（24 GB）上运行，并依据长叙事提示、每个主体的参考图像以及视觉定位框，生成时间连贯且保持身份一致的图像序列。三个协同模块驱动该系统：高斯中心注意力（GCA）动态聚焦于每个主体的核心区域并缓解视觉定位框重叠；动作增强奇异值重加权（AB-SVR）放大文本嵌入空间中与动作相关的方向；选择性遗忘缓存（SFC）保留可迁移的背景线索、遗忘非必要历史，并有选择地呈现所保留的线索以建立跨场景的语义联系。与基线方法相比，实验表明 CLIP-T 最高提升 10-15%，DreamSim 低于强基线，而 CLIP-I 保持在视觉上可接受且具竞争力的范围内。在 24 GB GPU 上以匹配的分辨率与步数运行时，推理速度快于 FluxKontext。在定性上，StoryTailor 呈现出富有表现力的交互以及不断演变却稳定的场景。
英文原文摘要 1258 字符
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 the retained cues to build cross scene semantic ties. Compared with baseline methods, the 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 a matched resolution and steps on a 24 GB GPU, inference is faster than FluxKontext. Qualitatively, StoryTailor delivers expressive interactions and evolving yet stable scenes.