Skip to content

Latest commit

 

History

History
67 lines (50 loc) · 2.75 KB

visualization.md

File metadata and controls

67 lines (50 loc) · 2.75 KB

Visualization tutorial

visualization for spd

In SPD, we utilize SUSTechPOINTS as the visualization tool for point cloud 3D bounding boxes and tracking IDS. SUSTechPOINTS is a Portable 3D Point Cloud Interactive Annotation Platform System. We offer a tool for converting labels from dair-v2x to SUS. For more details, please refer to gen_SUS_label

visualization for vic3d

We provide the tools to visualize the 3d label in images and point clouds, and visualize the prediction results.

visualize 3d label in image

example:

If you want to visualize the 3d label in images, you can run following commands.

cd ${dair_v2x_root}
python tools/visualize/vis_label_in_image.py --path ${data_root} --output-file ./vis_results
  • path refers to the directory of the images that want be visualized. It can be ${DAIR-V2X-V_ROOT}, ${DAIR-V2X-I_ROOT}, ${DAIR-V2X-C-I_ROOT}, or ${DAIR-V2X-C-V_ROOT}.
  • save_path refers to the path of the results of visualization.

visualize 3d label in point cloud

If you want to visualize the 3d label in point cloud, you can run following commands.

cd ${dair_v2x_root}
python tools/visualize/vis_label_in_3d.py --task pcd_label --pcd-path ${pcd_path} --label-path ${label_json_path}
  • --task refers to the type of task you choose to visualize, the optional values are 'fusion', 'single', 'pcd_label'. Here we should set the --task as 'pcd_label'.
  • --pcd-path refers to the the path of the pcd file that want be visualized.
  • --label-path refers to the the label path of the pcd file that want be visualized.

visualize 3d label with prediction results

example:

After evaluating the TCLF on VIC-Aync-2 dataset with following commands

cd ${dair_v2x_root}
cd v2x
bash scripts/eval_lidar_late_fusion_pointpillars.sh 0 late_fusion 2 0 100

you will generate following cache files including the model prediction results. The cache file structure should be like

└── cache
    ├───── tmps
    └───── vic-late-lidar
       ├───── inf
       ├───── veh
       └───── result

If you want to visualize the label and the VIC3D predictionn, you can run following commands.

cd ${dair_v2x_root}
python tools/visualize/vis_label_in_3d.py --task fusion --path v2x/cache/vic-late-lidar --id 0
  • --task refers to the type of task you choose to visualize, the optional values are 'fusion', 'single', 'pcd_label'.
  • --path refers to the pickle file generated during inference.
  • --id refers to the 'filename' you want to visualize.