用单步扩散模型实现逼真镜头虚化，解决深度误差导致的伪影问题。
BokehDiff: Neural Lens Blur with One-Step Diffusion
- 基于物理成像原理设计自注意力模块，融合景深与遮挡效应。
- 单步推理无额外噪声，生成结果质量高且保真度强。
- 用扩散模型合成透明前景，解决真实数据稀缺难题。
我们提出BokehDiff，一种新型镜头虚化渲染方法，借助生成式扩散先验实现物理准确且视觉美观的效果。以往方法受限于深度估计精度，在深度不连续处产生伪影。本方法采用符合图像形成过程的物理启发自注意力模块，引入依赖深度的模糊圈约束和自遮挡效应。将扩散模型适配至单步推理方案，无需引入额外噪声，实现高质量、高保真结果。针对缺乏可扩展成对数据的问题，提出用扩散模型合成具有透明度的逼真前景，兼顾真实性与场景多样性。
原文摘要 · Abstract (English)
We introduce BokehDiff, a novel lens blur rendering method that achieves physically accurate and visually appealing outcomes, with the help of generative diffusion prior. Previous methods are bounded by the accuracy of depth estimation, generating artifacts in depth discontinuities. Our method employs a physics-inspired self-attention module that aligns with the image formation process, incorporating depth-dependent circle of confusion constraint and self-occlusion effects. We adapt the diffusion model to the one-step inference scheme without introducing additional noise, and achieve results of high quality and fidelity. To address the lack of scalable paired data, we propose to synthesize photorealistic foregrounds with transparency with diffusion models, balancing authenticity and scene diversity.
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。