Visualizing 3D Bounding Box on Point Cloud

WeijingShi / Point-GNN

Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud, CVPR 2020.

MIT License

540 stars 112 forks source link

Visualizing 3D Bounding Box on Point Cloud #14

Closed amritpaul closed 4 years ago

amritpaul commented 4 years ago

Hi, Thanks a lot for providing the codebase. I have trained it and the results look pretty good as described in the paper. However, I wasn't able to figure out how to plot the predicted 3D bounding boxes on the point cloud. Can anyone help me resolve this issue?

WeijingShi commented 4 years ago

The run.py script contains a flag --level LEVEL to turn on some visualizations.

For example:

python3 run.py checkpoints/car_auto_T3_trainval/ --level 1 --test --dataset_root_dir DATASET_ROOT_DIR --output_dir DIR_TO_SAVE_RESULTS

This will enable a 3D view visualization with the boxes plotted. You can press [q] for the next frame. For consecutive visualization of without blocking the code, you can set --level 2.

I hope it helps.