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
150 stars 5 forks source link

Ablation for point_nsample #18

Closed GG-Bonds closed 5 days ago

GG-Bonds commented 2 weeks ago

I observed point_nsample==512 in the code. I would like to ask the author whether he has done ablation experiments on point_nsample. According to my understanding, the larger the point_nsample, the more supervision signals, and the effect should be better.

Nightmare-n commented 1 week ago

Due to limited GPU memory, we did not use a larger point_nsample to train the model. However, we experimented with point_nsample=256, and the results were not much different from point_nsample=512. I think the reason behind this is that during the training process, all points have a chance to be randomly sampled and seen by the network.