Official implementation of our TIV'23 paper: Occupancy-MAE: Self-supervised Pre-training Large-scale LiDAR Point Clouds with Masked Occupancy Autoencoders
I visualized training loss from tensorboard and it looks like it's too smooth and batch independent.
I'm not sure it's correct, would you like to share your insights ？
Hi, Min @chaytonmin
I visualized training loss from tensorboard and it looks like it's too smooth and batch independent. I'm not sure it's correct, would you like to share your insights ？