search

Nightmare-n / UniPAD

UniPAD: A Universal Pre-training Paradigm for Autonomous Driving (CVPR 2024)
https://arxiv.org/abs/2310.08370
Apache License 2.0
175 stars 7 forks source link

Visualization of rendered RGB&Depth images #20

Open BoynextDoor114514 opened 4 months ago

BoynextDoor114514 commented 4 months ago

Hi, thanks for this great work. I wanna know how to get rendered RGB&D-images as you've shown in your paper, could you plz give me some advice?

Nightmare-n commented 4 months ago

We save the rendered results at here.

BoynextDoor114514 commented 4 months ago

Thanks a lot for your reply, and I have one more question about the volumetric representation : since we can only train with Point Cloud or multi-view images, does that mean this volumetric representation cannot contain the features from two branches at the same time?
Since our aim is to get a pre-trained Point Cloud encoder for the downstream tasks, it's a little bit confusing for me to understand how it works.