DreamStyle：视频风格化的统一框架
DreamStyle: A Unified Framework for Video Stylization
DreamStyle 将文本、风格图像和风格化首帧三种条件统一到基于 I2V 扩散模型的 LoRA 微调框架中，其 token 特定的 up 矩阵设计可缓解多条件 token 混淆，为多模态条件控制生成提供了可复用的参数高效适配思路。
原文依据 trained using a Low-Rank Adaptation (LoRA) with token-specific up matrices that reduces the confusion among different condition tokens
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视频风格化作为视频生成模型的重要下游任务，尚未得到充分探索。其输入风格条件通常包括文本、风格图像和风格化首帧。每种条件都有其独特优势：文本更灵活，风格图像提供更准确的视觉锚点，风格化首帧使长视频风格化成为可能。然而，现有方法大多局限于单一类型的风格条件，限制了其应用范围。此外，缺乏高质量数据集导致风格不一致和时序闪烁。为解决这些局限，我们提出 DreamStyle，一个统一的视频风格化框架，支持（1）文本引导、（2）风格图像引导和（3）首帧引导的视频风格化，并配备精心设计的数据整理流程以获取高质量配对视频数据。DreamStyle 基于原始图像到视频（I2V）模型构建，并使用带有 token 特定 up 矩阵的低秩适配（LoRA）进行训练，以减少不同条件 token 之间的混淆。定性和定量评估均表明，DreamStyle 在所有三种视频风格化任务中均表现良好，并在风格一致性和视频质量方面优于对比方法。
英文原文摘要 1305 字符
Video stylization, an important downstream task of video generation models, has not yet been thoroughly explored. Its input style conditions typically include text, style image, and stylized first frame. Each condition has a characteristic advantage: text is more flexible, style image provides a more accurate visual anchor, and stylized first frame makes long-video stylization feasible. However, existing methods are largely confined to a single type of style condition, which limits their scope of application. Additionally, their lack of high-quality datasets leads to style inconsistency and temporal flicker. To address these limitations, we introduce DreamStyle, a unified framework for video stylization, supporting (1) text-guided, (2) style-image-guided, and (3) first-frame-guided video stylization, accompanied by a well-designed data curation pipeline to acquire high-quality paired video data. DreamStyle is built on a vanilla Image-to-Video (I2V) model and trained using a Low-Rank Adaptation (LoRA) with token-specific up matrices that reduces the confusion among different condition tokens. Both qualitative and quantitative evaluations demonstrate that DreamStyle is competent in all three video stylization tasks, and outperforms the competitors in style consistency and video quality.