search

sulaimanvesal / PointCloudUDA

[IEEE-TMI 2021] This is our PyTorch implementation for Adapt Everywhere paper on unsupervised domain adaptation using entropy and point-cloud paper.
MIT License
29 stars 3 forks source link

image input #4

Open wangm-ting opened 2 years ago

wangm-ting commented 2 years ago

It is a 2D point-cloud regression,why the input image is R( w x h x 3)?

sulaimanvesal commented 2 years ago

A point cloud is represented as a set of 3D points {Pi| i = 1, ..., n}, where each point Pi is a vector of its (x, y, z) coordinate plus extra feature channels such as color, normal etc. The input is 2D image but for each point cloud the (x, y, z) coordinates are required.