Awesome
Exploring Simple 3D Multi-Object Tracking for Autonomous Driving
<p align='left'> <img src='example.gif' width='721'/> </p>Chenxu Luo, Xiaodong Yang, Alan Yuille <br> Exploring Simple 3D Multi-Object Tracking for Autonomous Driving, ICCV 2021<br> [Paper] [Poster] [YouTube]
Getting Started
Installation
Please refer to INSTALL for the detail.
Data Preparation
python ./tools/create_data.py nuscenes_data_prep --root_path=NUSCENES_TRAINVAL_DATASET_ROOT --version="v1.0-trainval" --nsweeps=10
- Waymo Open Dataset (TODO)
Training
python -m torch.distributed.launch --nproc_per_node=8 ./tools/train.py examples/point_pillars/configs/nusc_all_pp_centernet_tracking.py --work_dir SAVE_DIR
Test
In ./model_zoo
we provide our trained (pillar based) model on nuScenes.
Note: We currently only support inference with a single GPU.
python ./tools/val_nusc_tracking.py examples/point_pillars/configs/nusc_all_pp_centernet_tracking.py --checkpoint CHECKPOINTFILE --work_dir SAVE_DIR
Citation
Please cite the following paper if this repo helps your research:
@InProceedings{Luo_2021_ICCV,
author = {Luo, Chenxu and Yang, Xiaodong and Yuille, Alan},
title = {Exploring Simple 3D Multi-Object Tracking for Autonomous Driving},
booktitle = {International Conference on Computer Vision (ICCV)},
year = {2021}
}
License
Copyright (C) 2021 QCraft. All rights reserved. Licensed under the CC BY-NC-SA 4.0 (Attribution-NonCommercial-ShareAlike 4.0 International). The code is released for academic research use only. For commercial use, please contact business@qcraft.ai.