This repository contains the code (in Keras) for "Fast and Robust UAV to UAV Detection and Tracking" by Jing Li.

If you use our software or our data sets, please cite:

J. Li, D. Ye, M. Kolsch, J. Wachs and C. Bouman, "Fast and Robust UAV to UAV Detection and Tracking from Video" in IEEE Transactions on Emerging Topics in Computing. doi: 10.1109/TETC.2021.3104555 url: https://doi.ieeecomputersociety.org/10.1109/TETC.2021.3104555

@ARTICLE {9519550, author = {J. Li and D. Ye and M. Kolsch and J. P. Wachs and C. A. Bouman}, journal = {IEEE Transactions on Emerging Topics in Computing}, title = {Fast and Robust UAV to UAV Detection and Tracking from Video}, year = {5555}, volume = {}, number = {01}, issn = {2168-6750}, pages = {1-1}, keywords = {target tracking;cameras;detectors;unmanned aerial vehicles;optical imaging;radar tracking;optical detectors}, doi = {10.1109/TETC.2021.3104555}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, month = {aug} }

https://www.computer.org/csdl/journal/ec/5555/01/09519550/1wc8Vbe1r7G

@inproceedings{li2016multi,
title={Multi-target detection and tracking from a single camera in Unmanned Aerial Vehicles (UAVs)},
author={Li, Jing and Ye, Dong Hye and Chung, Timothy and Kolsch, Mathias and Wachs, Juan and Bouman, Charles},
booktitle={2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
pages={4992--4997},
year={2016},
organization={IEEE} }

1. Requirements
2. Usage
3. Results
4. Contacts

anaconda
CUDA >= 10.0
CUDA Driver Version: >=430

run "demo.sh" to test the example video

Saved under Experiment_Results folder:
.txt file with detected bounding boxes,
videos with groundtruth and detection results

https://engineering.purdue.edu/~bouman/UAV_Dataset/

[email protected]

Any discussions or concerns are welcomed!