Skip to content
Branch: master
Find file History
Pull request Compare This branch is 31 commits ahead of ApolloScapeAuto:master.
Latest commit a6a76c3 Jul 20, 2019
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
eval.py add detection tracking eval code Jun 27, 2019
readme.md
run.sh add detection tracking eval code Jun 27, 2019
utils.py

readme.md

The 3D Lidar Object Detection and Tracking Challenge of Apolloscape Dataset

For detail and download

Introduction

Our 3D Lidar object detection and tracking dataset consists of LiDAR scanned point clouds with high quality annotation. It is collected under various lighting conditions and traffic densities in Beijing, China. More specifically, it contains highly complicated traffic flows mixed with vehicles, cyclists, and pedestrians.

Evaluation

eval.py is the evaluation code. Run the code for a sample evaluation:

source activate apolloscape

# export NUMBA_ENABLE_CUDASIM=1
# export CUDA_VISIBLE_DEVICES=4
# export CUDA_VISIBLE_DEVICES=4,5,6,7

# tracking
python eval.py --modeType=tracking --gtPath=../track/apollo_lab --dtPath=../track/apollo_res --typeFilterFlag
# detection
# python eval.py --gtPath=apollo_lab_test --dtPath=apollo_res_test --apSampleNum=10 #--typeFilterFlag #2>&1 | tee run.log

Submission of data format

Submit your result for online evaluation here: Submit

Leaderboard

Contact

Please feel free to contact us, or raise an issue with any questions, suggestions or comments:

You can’t perform that action at this time.