A 3D object detector for the NuScenes dataset. Designed to work with a hardware monocular object detector and radar detections. In this work, camera detection points are generated with the CenterTrack algorithm, but this could be done using any 3D object detector.
- cmake >=3.17
- pybind11 >=2.2 (installed with apt is easier)
- CUDA (tested on 10.2 & 11.2)
Installation is simplified with python.
Simply run python setup.py install
and dependencies will be installed and configured.
-
Follow the install steps above to setup the python library.
-
Run CenterTrack on all splits of the nuscenes dataset.
These will be pickled together to allow for faster loading later. If you don't need all of them (for example, only the test for evaluation, then you can only run the test set)
Here's the CenterTrack data I prepared before! note: It does not include the full dataset. Only the mini and test sets. Extract this to the
results/CenterTrack
directory. -
From the root of this repository, run
./pySrc/main.py [nuscenes_version] [nuscenes_split] [nuscenes_root]
A couple of changes have been made to centertrack to allow for it to run on NuScenes.
These changes can be found here.
This requires PyTorch version 1.4.0, torchvision 0.5.0, and Cuda <= 10.2
-
Follow the setup steps in the centertrack repository
These are mostly in readme/INSTALL.md
-
Run the
convert_nuscenes
script. You'll have to adjust the DATA_ROOT at the top of this file -
Create an alias in the root of your nuscenes directory called
anotations
that points to the directory containing the.json
files just created. -
Download the pretrained model called
nuScenes_3Ddetection_e140
listed inreadme/MODEL_ZOO.md
. Place this in themodels
folder. -
Run the following script, adjusting
dataset_version
for each split of the dataset you needpython test.py ddd --exp_id nusc_det_full --load_model ../models/nuScenes_3Ddetection_e140.pth --dataset nuscenes --dataset_version mini-train
and copy the
.json
file with the results in theexp/ddd/$exp_id$
folder to one ofresults/CenterTrack/{train,val,mini-train,mini-val,test}
.