OpenIllumination

OpenIllumination数据集由加州大学圣地亚哥分校等机构联合创建，包含超过108,000张真实世界物体的图像，涵盖64个具有多样材料的物体，通过72个不同的摄像机视角和大量不同照明条件捕捉。数据集为每个图像提供了精确的摄像机参数、照明地面实况和前景分割掩码，旨在为真实物体的逆向渲染和材料分解方法提供定量评估。该数据集适用于多种逆向渲染任务，如物体几何、材料和照明的恢复，以及图像重照明和新视图合成等应用。

The OpenIllumination dataset was jointly created by the University of California, San Diego and other institutions. It contains over 108,000 images of real-world objects, covering 64 objects with diverse materials, which were captured from 72 distinct camera viewpoints and under a wide range of illumination conditions. The dataset provides precise camera parameters, illumination ground truth, and foreground segmentation masks for each image. It aims to provide quantitative evaluation for inverse rendering and material decomposition methods of real-world objects. This dataset is suitable for a variety of inverse rendering tasks, such as the recovery of object geometry, materials and illumination, as well as applications like image relighting and novel view synthesis.