P = ((S-1)*W-S+F)/2, with F = filter size, S = stride

`ARKitScenes Dataset class. Args: root (string): Root directory of dataset where directory exists or will be saved to if down

Taken from torchvision - slightly adjusted

Inspired by: Depth Map Super-Resolution by Deep Multi-Scale Guidance http://personal.ie.cuhk.edu.hk/~ccloy/files/eccv_2016_depth.pdf

Inspired by: Multi-Scale Progressive Fusion Learning for Depth Map Super-Resolution https://arxiv.org/pdf/2011.11865v1.pdf variables:

r"""Samples elements randomly over multiple epochs Arguments: data_source (Dataset): dataset to sample from num_iter (int) : Numb

Inspired by: Multi-Scale Progressive Fusion Learning for Depth Map Super-Resolution https://arxiv.org/pdf/2011.11865v1.pdf

Inspired by: Multi-Scale Progressive Fusion Learning for Depth Map Super-Resolution https://arxiv.org/pdf/2011.11865v1.pdf

Calculating Average Precision