Open Vincento-Wang opened 2 months ago
Hi, from which paper is this screenshot taken? In any case, as described in section 5.1 of the NeRSemble paper, we use BackgroundMattingV2 to obtain high-quality alpha maps for each image. BackgroundMattingV2 requires an additional image of the empty background which we also provide as part of the NeRSemble dataset.
Thank you for your great work, I wonder how to get the mask of gt for rest of the dataset.