arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2503.16129cs.GR
收藏

基于可控分割的文本引导风格编辑

Controllable Segmentation-Based Text-Guided Style Editing

Jingwen Li, Aravind Chandrasekar, Mariana Rocha, Chao Li, Yuqing Chen

AI总结:

本研究提出一种基于StyleMamba框架、结合语义分割模型的文本引导区域可控风格编辑方法,通过区域条件向量与区域特定方向损失实现高保真转换,在真实复杂场景风格化中优于纯全局风格迁移方法。

摘要:

我们提出了一种由文本提示驱动的可控、区域特定风格编辑的新方法。本方法以StyleMamba(状态空间风格对齐框架)为基础,将语义分割模型集成到风格迁移流程中。用户可选择性地将文本驱动的风格变化应用于特定分割区域(例如“将建筑变成赛博朋克塔楼”),同时保持其他区域(例如“人物”或“树木”)不变。通过引入区域条件向量和区域特定方向损失,本方法实现了高保真转换,既符合语义边界,又匹配用户指定的风格描述。大量实验表明,我们的方法能灵活处理真实场景中的复杂场景风格化任务,在可控性和质量上均优于纯全局风格迁移方法。

英文摘要:

We present a novel approach for controllable, region-specific style editing driven by textual prompts. Building upon the state-space style alignment framework introduced by \emph{StyleMamba}, our method integrates a semantic segmentation model into the style transfer pipeline. This allows users to selectively apply text-driven style changes to specific segments (e.g., ``turn the building into a cyberpunk tower'') while leaving other regions (e.g., ``people'' or ``trees'') unchanged. By incorporating region-wise condition vectors and a region-specific directional loss, our method achieves high-fidelity transformations that respect both semantic boundaries and user-driven style descriptions. Extensive experiments demonstrate that our approach can flexibly handle complex scene stylizations in real-world scenarios, improving control and quality over purely global style transfer methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑