基于可微光线追踪的镜片组件删除
Lens Component Deletion based on Differentiable Ray Tracing
把可微光线追踪与后处理网络端到端联合优化，并给出基于 Rayleigh-Sommerfeld 衍射模型的高效 PSF 估计，可直接用于计算摄影中镜头-算法协同设计、像差校正与轻量化成像系统的搭建。
原文依据 we propose a novel lens component deletion pipeline for miniature optical systems, which automatically deletes the suitable lens component, and then optimizes both the lens system and the post-processing network
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论为实现光学镜头系统的紧凑化或降低成本，设计者通常依赖商业软件独立设计镜头系统，与后处理算法相互割裂，导致对设计者专业知识的过度依赖，且往往需要耗费大量时间。近期，利用可微光线追踪的联合优化方法相继出现，在镜头设计任务中展现出显著潜力。然而，现有流程对微型光学系统缺乏准确且高效的衍射建模能力。在本工作中，我们提出一种面向微型光学系统的新型镜片组件删除流程，该流程自动删除合适的镜片组件，随后对镜头系统与后处理网络进行联合优化，以实现联合像差校正。此外，我们引入一种新的指标，用于评估光学系统中每个镜片组件的贡献，旨在识别对系统影响最小的镜片组件。我们还基于 Rayleigh-Sommerfeld 衍射模型，开发了一种高效的可微点扩散函数估计方法，显著降低了 GPU 显存消耗。我们提出的流程在删除镜片组件的同时，保持与原始镜头系统相当的成像质量，从而实现微型光学系统的紧凑化或低成本优化。项目主页：https://github.com/WenguanZhang/HappyLens
英文原文摘要 1478 字符
To achieve compactness or cost reduction for optical lens systems, designers typically rely on commercial software to design lens systems independently of post-processing algorithms, leading to excessive dependence on designers' expertise and often requiring significant time. Recently, joint optimization approaches utilizing differentiable ray tracing have emerged, demonstrating significant potential in lens design tasks. However, existing pipelines lack accurate and efficient diffraction modeling capabilities for miniature optical systems. In this work, we propose a novel lens component deletion pipeline for miniature optical systems, which automatically deletes the suitable lens component, and then optimizes both the lens system and the post-processing network to achieve joint aberration correction. Additionally, we introduce a novel metric for evaluating the contribution of each lens component within an optical system, aimed at identifying the lens component that has the least impact on the system. We also develop an efficient differentiable point spread function estimation method based on the Rayleigh-Sommerfeld diffraction model, significantly reducing GPU memory consumption. Our proposed pipeline achieves lens component deletion while maintaining imaging quality comparable to the original lens system, thereby enabling the compactness or cost-effective optimization of miniature optical systems. Project page: https://github.com/WenguanZhang/HappyLens