Learning 3D Representations from 2D Pre-Trained Models via Image-to-Point Masked Autoencoders.

Renrui Zhang, Liuhui Wang, Yu Qiao, Peng Gao, Hongsheng Li

Browse the full CVPR paper archive.