基于扩散框架实现照片级真实且高效的散景渲染
Towards Photorealistic and Efficient Bokeh Rendering via Diffusion Framework
将散景渲染与超分辨率统一进单个扩散框架，并用焦点感知掩码注意力与退化感知深度模块处理低分辨率真实图像，为手机端计算摄影中「先复原再渲染」的两阶段流水线提供了一体化替代方案。
原文依据 our method jointly optimizes bokeh rendering and super-resolution
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现有移动设备受限于紧凑的光学设计（如小孔径），难以产生自然、符合光学真实的散景效果。尽管近期的基于学习的方法已展现出良好结果，但它们在高数码变焦下拍摄的照片上仍表现不佳，这类照片往往分辨率下降且细节丢失。一种朴素的解决方案是在进行散景渲染之前先增强图像质量，但这种两阶段流水线降低了效率并引入不必要的误差累积。为克服这些局限，我们提出 MagicBokeh，一个面向高质量、高效散景渲染的统一扩散框架。通过交替训练策略与焦点感知掩码注意力机制，本方法联合优化散景渲染与超分辨率，显著提升了可控性与视觉保真度。此外，我们引入退化感知深度模块，以便从低质量输入中获得更准确的深度估计。实验结果表明，MagicBokeh 能够高效地产生照片级真实的散景效果，尤其是在真实世界的低分辨率图像上，为散景渲染的未来发展铺平了道路。我们的代码与模型已发布于 this \href https://github.com/vivoCameraResearch/MagicBokeh url 。
英文原文摘要 1356 字符
Existing mobile devices are constrained by compact optical designs, such as small apertures, which make it difficult to produce natural, optically realistic bokeh effects. Although recent learning-based methods have shown promising results, they still struggle with photos captured under high digital zoom levels, which often suffer from reduced resolution and loss of fine details. A naive solution is to enhance image quality before applying bokeh rendering, yet this two-stage pipeline reduces efficiency and introduces unnecessary error accumulation. To overcome these limitations, we propose MagicBokeh, a unified diffusion-based framework designed for high-quality and efficient bokeh rendering. Through an alternative training strategy and a focus-aware masked attention mechanism, our method jointly optimizes bokeh rendering and super-resolution, substantially improving both controllability and visual fidelity. Furthermore, we introduce degradation-aware depth module to enable more accurate depth estimation from low-quality inputs. Experimental results demonstrate that MagicBokeh efficiently produces photorealistic bokeh effects, particularly on real-world low-resolution images, paving the way for future advancements in bokeh rendering. Our code and models are available at this \href https://github.com/vivoCameraResearch/MagicBokeh url .