dwofk / fast-depth

ICRA 2019 "FastDepth: Fast Monocular Depth Estimation on Embedded Systems"
MIT License
926 stars 189 forks source link

train transform #54

Open sunmengnan opened 3 years ago

sunmengnan commented 3 years ago

in the first line of train transform, why is transforms.Resize(250.0 / iheight) for computational efficiency? transform = transforms.Compose([ transforms.Resize(250.0 / iheight), # this is for computational efficiency, since rotation can be slow transforms.Rotate(angle), transforms.Resize(s), transforms.CenterCrop((228, 304)), transforms.HorizontalFlip(do_flip), transforms.Resize(self.output_size), ])