Depth map and 3D reconstruction

lppllppl920 / EndoscopyDepthEstimation-Pytorch

Official Repo for the paper "Dense Depth Estimation in Monocular Endoscopy with Self-supervised Learning Methods" (TMI)

GNU General Public License v3.0

124 stars 27 forks source link

Depth map and 3D reconstruction #12

Closed caozhengtao110 closed 3 years ago

caozhengtao110 commented 3 years ago

hello! How can I get the final depth map and 3D reconstruction results from the data you gave me? I have tried to adjust the training phase, will the test phase get the final result? Looking forward to your reply! I hope to get your help.

lppllppl920 commented 3 years ago

The 3D reconstruction result needs to be obtained by converting the dense depth map predicted from the network to a 3D point cloud. It requires camera intrinsic matrix as an additional input.