用单张RGB图生成逼真偏振图像，无需3D资产
PolarAnything: Diffusion-based Polarimetric Image Synthesis
- 基于扩散模型，仅凭一张RGB图生成偏振图像
- 生成图像保真度高，支持形状恢复等下游任务
- 适合缺乏偏振相机的科研与工业场景
偏振图像有助于图像增强与三维重建，但偏振相机获取困难，限制了其广泛应用。为此，亟需合成逼真偏振图像。现有模拟器Mitsuba依赖参数化偏振成像模型，并需覆盖形状与真实物理渲染（PBR）材质的大量3D资产，难以生成大规模逼真图像。为此，我们提出PolarAnything，仅需单张RGB输入即可生成兼具真实感与物理准确性的偏振图像，摆脱对3D资产库的依赖。受预训练扩散模型零样本能力启发，我们设计了一种基于扩散的生成框架，结合有效表征策略，保持偏振属性保真度。实验表明，该模型可生成高质量偏振图像，并支持从偏振中恢复形状等下游任务。
原文摘要 · Abstract (English)
Polarization images facilitate image enhancement and 3D reconstruction tasks, but the limited accessibility of polarization cameras hinders their broader application. This gap drives the need for synthesizing photorealistic polarization images. The existing polarization simulator Mitsuba relies on a parametric polarization image formation model and requires extensive 3D assets covering shape and PBR materials, preventing it from generating large-scale photorealistic images. To address this problem, we propose PolarAnything, capable of synthesizing polarization images from a single RGB input with both photorealism and physical accuracy, eliminating the dependency on 3D asset collections. Drawing inspiration from the zero-shot performance of pretrained diffusion models, we introduce a diffusion-based generative framework with an effective representation strategy that preserves the fidelity of polarization properties. Experiments show that our model generates high-quality polarization images and supports downstream tasks like shape from polarization.
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。