INTEL-TAU

这个名为 INTEL-TAU 的数据集共有 7022 个场景，是用于照明估计研究的最大可用数据集。使用三种不同相机模型（即佳能 5DSR、尼康 D810 和索尼 IMX135）捕获的各种场景使得数据集适合评估不同照明估计技术的相机和场景不变性。隐私屏蔽是针对敏感信息（例如人脸）进行的。因此，它与通用数据保护条例 (GDPR) 条例一致。此外，可以使用 INTEL-TAU 评估移动图像的颜色阴影效果，因为我们提供原始数据的校正和未校正版本。

The dataset named INTEL-TAU, comprising 7022 distinct scenes, is the largest publicly available dataset for illumination estimation research. Captured with three different camera models (i.e., Canon 5DSR, Nikon D810, and Sony IMX135), its diverse scenes allow for the evaluation of camera and scene invariance across various illumination estimation technologies. Sensitive information such as human faces has undergone privacy masking, rendering the dataset compliant with the General Data Protection Regulation (GDPR). Furthermore, INTEL-TAU can be employed to evaluate color shading artifacts in mobile images, as both corrected and uncorrected versions of the raw data are provided.