Pretrain, Self-train, Distill: A simple recipe for Supersizing 3D Reconstruction.

Kalyan Vasudev Alwala Abhinav Gupta Shubham Tulsiani

Published in: CVPR (2022)

Keyphrases

real time
structure from motion
highly reliable
shape recovery
machine learning
learning algorithm
clustering algorithm
three dimensional
feature space
probabilistic model
d scene
urban scenes