Eaphan / Robust3DOD

A Comprehensive Study of the Robustness for LiDAR-based 3D Object Detectors against Adversarial Attacks
https://arxiv.org/abs/2212.10230
Apache License 2.0
92 stars 4 forks source link

Visualization #3

Open lesmiserable opened 3 months ago

lesmiserable commented 3 months ago

Hello, thank you for sharing. The appendix at the end of the paper visualizes a bird's-eye view of the point cloud and its 3D detection box and ground-truth bounding box (as below). I would like to ask how this visualization result is generated? Thanks again. 1721306467795

Eaphan commented 2 months ago

We often use open3d or mayavi to visualize the point cloud and boxes. Here are some examples:

https://github.com/Eaphan/STEMD/blob/main/datasets/visualization/waymo_vis/point_cloud_3D_vis_with_boxes.py https://github.com/Eaphan/STEMD/blob/main/datasets/visualization/waymo_vis/point_cloud_BEV_vis_with_boxes.py

lesmiserable commented 2 months ago

Thanks a lot, I will try to use it on the KITTI dataset.

Eaphan commented 2 months ago

You are welcome!