pip install nyu-depth-v2
from torchvision import transforms
from nyuv2.NYUv2Dataset import NYUv2Dataset
transform_image = transforms.Compose([ ... ])
transform_depth = transforms.Compose([ ... ])
dataset = NYUv2Dataset(
"./dataset",
True,
transform_image,
transform_depth)
This requires torch
, torchvision
, einops
, numpy
, h5py
, opencv-python
, pathlib
.
python310