The image size used for depth evaluation

[flamehaze1115]

Hello! I found that the evaluation results of DPSNET on SUN3D in your paper have obvious differences with the original paper. I check your code, and I found that you downsample the GT depth to (320,240) to evaluate with pred_depth while prior works usually upsample the pred_depth to the original size to do the evaluation. Although this minor difference won't influence your paper's conclusion, maybe you could clarify this difference in the READE.ME to prevent other readers from misusing your numbers reported in your paper. If possible, may you report your tables with the original size like other works and readers could easily cite your tables?

Thank you.

[sinhaayan87]

Thanks for pointing this out. First author here. I am no longer affiliated with Magic Leapvband hence cannot modify the readme, or update the paper in any form. The image resolution has been stated in the main manuscript. There was no scale based augmentation during training, and likely that the network will need to be fine-tuned to directly work on the original image resolution. This code-base can be used to predict it at lower resolution and upsampled to desired resolution as you mentioned.