by Zhengqi Li,Noah SnavelyUnknown
by Zhengqi Li,Noah SnavelyLicense : Unknown
CGINTRINSICS is a new, large-scale dataset of physically-based rendered images of scenes with full ground truth decompositions. It includes > 20,000 high quality rendered images from Mitsuba Renderer. End-to-end training can be applied on the dataset to learn better scene decompositions.