用可空间自适应的4维查表实现逼真风格迁移，兼顾色彩精准与结构完整。
SA-LUT: Spatial Adaptive 4D Look-Up Table for Photorealistic Style Transfer
- 设计可学习的4D查表生成器，结合风格图像多尺度特征动态生成颜色映射。
- 引入内容-风格交叉注意力生成上下文图，实现局部像素级颜色调整。
- 在新基准PST50上超越现有方法，视频处理达16帧/秒，实时高效。
逼真风格迁移（PST）通过适配参考图像色彩来实现真实世界的调色效果，同时保持内容结构不变。现有方法主要分为两类：生成式方法虽注重风格保真度，但牺牲内容完整性和效率；全局色彩变换方法如3D LUT能保持结构，却缺乏局部适应性。为此，本文提出空间自适应4维查表（SA-LUT），融合LUT的高效性与神经网络的可适应性。SA-LUT包含两个核心模块：(1) 风格引导的4D LUT生成器，从风格图像提取多尺度特征以预测4D LUT；(2) 上下文生成器，利用内容-风格交叉注意力生成上下文图，实现空间自适应调整，使4D LUT能在保持结构完整性的同时进行精确颜色变换。为建立严谨的评估体系，我们提出首个专用于PST评估的基准PST50。实验表明，相比3D LUT方法，SA-LUT在LPIPS指标上降低66.7%，并实现16 FPS的视频风格化实时性能。代码与数据集已开源。
原文摘要 · Abstract (English)
Photorealistic style transfer (PST) enables real-world color grading by adapting reference image colors while preserving content structure. Existing methods mainly follow either approaches: generation-based methods that prioritize stylistic fidelity at the cost of content integrity and efficiency, or global color transformation methods such as LUT, which preserve structure but lack local adaptability. To bridge this gap, we propose Spatial Adaptive 4D Look-Up Table (SA-LUT), combining LUT efficiency with neural network adaptability. SA-LUT features: (1) a Style-guided 4D LUT Generator that extracts multi-scale features from the style image to predict a 4D LUT, and (2) a Context Generator using content-style cross-attention to produce a context map. This context map enables spatially-adaptive adjustments, allowing our 4D LUT to apply precise color transformations while preserving structural integrity. To establish a rigorous evaluation framework for photorealistic style transfer, we introduce PST50, the first benchmark specifically designed for PST assessment. Experiments demonstrate that SA-LUT substantially outperforms state-of-the-art methods, achieving a 66.7% reduction in LPIPS score compared to 3D LUT approaches, while maintaining real-time performance at 16 FPS for video stylization. Our code and benchmark are available at https://github.com/Ry3nG/SA-LUT
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。