the issue regarding feature dimensions

ZrrSkywalker / I2P-MAE

[CVPR 2023] Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders

219 stars 17 forks source link

Closed yuhbai closed 10 months ago

yuhbai commented 10 months ago

Using the I2P_MAE.py extracts the features, why the output dimension of the second stage is (32, 255, 192) rather than (32, 256, 192)