VAD icon indicating copy to clipboard operation
VAD copied to clipboard

[ICCV 2023] VAD: Vectorized Scene Representation for Efficient Autonomous Driving

Results 50 VAD issues
Sort by recently updated
recently updated
newest added

Hi, and thank you for your work. When reviewing your evaluation code I find that your computation of the L2 displacement error ([here](https://github.com/hustvl/VAD/blob/1ede800a517653837faf78b5ec1c3b5fc3ec77b5/projects/mmdet3d_plugin/VAD/VAD.py#L624C1-L629C1)) is computed using the average displacement error...

Hi, I have few questions regarding Carla results. 1) For the paper, have you trained VAD on Carla and then reported the evaluation results or used your open loop checkpoint...

When will VADv2 be open sourced?

Hi, Thanks for your great work and contribution to the community. I noticed that [here](https://github.com/hustvl/VAD/blob/main/projects/mmdet3d_plugin/bevformer/modules/transformer.py#L158) and [here](https://github.com/hustvl/VAD/blob/main/projects/configs/VAD/VAD_base_e2e.py#L170) in your code are actually using ego-states from can-bus as initialized queries in...

Hello, Thank you for your great work! Below is my reproduced results, I found _plan_L2_ is the same with paper's, but the _plan_obj_box_col_ is not, I don't konw why. Do...

Hello, thank you for your great work! I find you write below comment in your paper: _‡ denotes taking ego status information as input._ So I want to konw whether...

Hi, Thank you for your work. I was wondering if you are aware of any efforts in training/validating your work on other datasets e.g., Argoverse2, Waymo, or ZOD?

Why does Agent Query directly perform motion transformer without performing tracking tasks? Will this reduce performance?

I found that the setting "rotate_prev_bev" in config is always set to True, which means the prev bev feat needs to be rotated. But the rotate_center is always equal to...

i find the batch fix to 1, what is the reason?