Home

Awesome

IoUattack

:herb: IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking

Shuai Jia, Yibing Song, Chao Ma and Xiaokang Yang

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021.

Introduction

<img src="https://github.com/VISION-SJTU/IoUattack/blob/main/demo/intro.png" width='500'/><br/>

We observe that the increase of noise level positively correlates to the decrease of IoU scores, but their directions are not exactly the same.

Results

Result for SiamRPN++ on multiple datasets

VOT2019<br>A / R / EAOVOT2018<br>A / R / EAOVOT2016<br>A / R / EAOVOT2018lt <br> F-scoreOTB2015<br>OP / DPNFS30<br>OP / DP
SiamRPN++0.596 / 0.472 / 0.2870.602 / 0.239 / 0.4130.643 / 0.200 / 0.4610.6250.695 / 0.9050.509 / 0.601
SiamRPN++(Random)0.591 / 0.727 / 0.2200.587 / 0.365 / 0.3010.632 / 0.340 / 0.3310.5530.631 / 0.8180.466 / 0.550
SiamRPN++(Attack)0.575 / 1.575 / 0.1240.568 / 1.171 / 0.1290.605 / 0.802 / 0.1830.4530.499 / 0.6440.394 / 0.446

Result for DiMP on multiple datasets

VOT2019<br>A / R / EAOVOT2018<br>A / R / EAOVOT2016<br>A / R / EAOVOT2018lt <br> F-scoreOTB2015<br>OP / DPNFS30<br>OP / DP
DiMP0.568 / 0.277 / 0.3320.574 / 0.145 / 0.4270.599 / 0.140 / 0.4490.6090.671 / 0.8690.614 / 0.729
DiMP(Random)0.567 / 0.373 / 0.2840.560 / 0.202 / 0.3630.592 / 0.168 / 0.4040.5550.659 / 0.8600.591 / 0.710
DiMP(Attack)0.474 / 0.641 / 0.1950.507 / 0.400 / 0.2480.536 / 0.374 / 0.2560.4430.592 / 0.7910.545 / 0.658

Result for LTMU on multiple datasets

VOT2019<br>A / R / EAOVOT2018<br>A / R / EAOVOT2016<br>A / R / EAOVOT2018ltT <br> F-scoreOTB2015<br>OP / DPNFS30<br>OP / DP
LTMU0.625 / 0.913 / 0.2010.624 / 0.702 / 0.1950.661 / 0.522 / 0.2360.6910.672 / 0.8720.631 / 0.764
LTMU(Random)0.623 / 1.073 / 0.1750.622 / 0.805 / 0.1780.646 / 0.592 / 0.2330.6570.622 / 0.8150.579 / 0.699
LTMU(Attack)0.576 / 1.470 / 0.1500.590 / 1.320 / 0.1200.604 / 0.904 / 0.1700.5890.517 / 0.7120.462 / 0.559

:herb: All raw results are available. [Google_drive]

Code

:herb: The code of IoU attack for SiamRPN++ is released!!

Test the original performance on VOT2018 dataset, please use the following command.

cd pysot/experiments/siamrpn_r50_l234_dwxcorr
python -u ../../tools/test_original.py 	\
	--snapshot model.pth 	\ # model path
	--dataset VOT2018 	\ # dataset name
	--config config.yaml	  # config file

Test IoU attack on VOT2018 dataset, please use the following command.

cd pysot/experiments/siamrpn_r50_l234_dwxcorr
python -u ../../tools/test_IoU_attack.py 	\
	--snapshot model.pth 	\ # model path
	--dataset VOT2018 	\ # dataset name
	--config config.yaml	  # config file

For the adversarial attack of other datasets, you should change the dataset name as mentioned above.

Demo

<img src="https://github.com/VISION-SJTU/IoUattack/blob/main/demo/car_clean.gif" width='300'/> <img src="https://github.com/VISION-SJTU/IoUattack/blob/main/demo/car_attack.gif" width='300'/><br/>         <img src="https://github.com/VISION-SJTU/IoUattack/blob/main/demo/legend.png" width='300'/><br/>

Citation

If any part of our paper and code is helpful to your work, please generously citing:

@inproceedings{jia-cvpr21-iouattack,
  title={IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking},
  author={Jia, Shuai and Song, Yibing and Ma, Chao and Yang, Xiaokang},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2021}
}

Thank you :)

Reference

We choose three representative trackers, SiamRPN++, DiMP and LTMU. The original code of these trackers are list as follows:

We also refer to the code of Boundary Attack for IoU attack.

Thanks for their wonderful works!

License

Licensed under an MIT license.