search

hrzhou2 / refinenet

7 stars 2 forks source link

NYUv2 Questions #3

Open kantai12138 opened 3 months ago

kantai12138 commented 3 months ago

Hello! I would like to ask how the results of NYUv2 can be converted into images? And how do you train the data of NYUv2? This is clearly an image rather than a point cloud. Looking forward to your reply. Thank you!

hrzhou2 commented 3 months ago

NYUv2 is a RGB-D dataset, and we use its depth estimation as the input point cloud.

kantai12138 commented 3 months ago

And could you provide your script for converting point clouds to depth maps for NYU?