Open VitorGuizilini-TRI opened 4 years ago
Hi, one quick question about the NYUv2 dataset: on the paper it is mentioned that the training split contains 24231 image-depth pairs, however the script provided in this repository only produces 1449 pairs. Which one is correct?
Hi, one quick question about the NYUv2 dataset: on the paper it is mentioned that the training split contains 24231 image-depth pairs, however the script provided in this repository only produces 1449 pairs. Which one is correct?