Giters

OpenDriveLab / PPGeo

[ICLR 2023] Pytorch implementation of PPGeo, a fully self-supervised driving policy pre-training framework to learn from unlabeled driving videos.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

编码器问题

Bobo-y opened this issue · comments

commented

实验中第一阶段用到的depth和pose 编码器是resnet 结构，第二阶段motionet 编码器也是resnet 结构，二者的设定有相关性吗？我理解第二阶段的视觉编码器应该是根据自己需求选用的吧？thanks

commented

Yes, you can use other vision encoder structures in the second stage.