Journal of the Korea Institute of Military Science and Technology (한국군사과학기술학회지)

The Korea Institute of Military Science and Technology (한국군사과학기술학회)
권호 표지
DOI QR코드

DOI QR Code

Occlusion Robust Military Vehicle Detection using Two-Stage Part Attention Networks

2단계 부분 어텐션 네트워크를 이용한 가려짐에 강인한 군용 차량 검출

  • Cho, Sunyoung (Defense AI Technology Center, Agency for Defense Development)
  • 조선영 (국방과학연구소 국방인공지능기술센터)
  • Received : 2022.01.04
  • Accepted : 2022.06.29
  • Published : 2022.08.05
Download PDF
⟨ Previous Next ⟩

Abstract

Detecting partially occluded objects is difficult due to the appearances and shapes of occluders are highly variable. These variabilities lead to challenges of localizing accurate bounding box or classifying objects with visible object parts. To address these problems, we propose a two-stage part-based attention approach for robust object detection under partial occlusion. First, our part attention network(PAN) captures the important object parts and then it is used to generate weighted object features. Based on the weighted features, the re-weighted object features are produced by our reinforced PAN(RPAN). Experiments are performed on our collected military vehicle dataset and synthetic occlusion dataset. Our method outperforms the baselines and demonstrates the robustness of detecting objects under partial occlusion.

Keywords

References

  1. H. Zhu, P. Tang, J. Park, S. Park, A. Yuille, "Robustness of Object Recognition Under Extreme Occlusion in Humans and Computational Models," CoRR, Vol. abs/1905.04598, 2019.
  2. A. Fawzi, P. Frossard, "Measuring the Effect of Nuisance Variables on Classifiers," BMVC, 2016.
  3. T. DeVries, G. W. Taylor, "Improved Regularization of Convolutional Neural Networks with Cutout," arXiv preprint arXiv:1708.04552, 2017.
  4. S. Yun, D. Han, S. J. Oh, S. Chun, Y. Yoo, "Cutmix: Regularization Strategy to Train Strong Classifiers with Localizable Features," ICCV, pp. 6023-6032, 2019.
  5. C. Zhou, J. Yuan, "Multi-Label Learning of Part Detectors for Heavily Occluded Pedestrian Detection," ICCV, pp. 3486-3495, 2017.
  6. C. Zhou, J. Yuan, "Non-Rectangular Part Discovery for Object Detection," BMCV, 2014.
  7. Y. Tian, P. Luo, X. Wang, X. Tang, "Deep Learning Strong Parts for Pedestrian Detection," ICCV, pp. 1904-1912, 2015.
  8. C. Zhou, J. Yuan, "Occlusion Pattern Discovery for Object Detection and Occlusion Reasoning," TCSVT, Vol. 30, No. 7, pp. 2067-2080, 2020.
  9. C. Zhou, J. Yuan, "Bi-Box Regression for Pedestrian Detection and Occlusion Estimation," ECCV, pp. 135-151, 2018.
  10. S. Yan, Q. Liu, "Inferring Occluded Features for Fast Object Detection," Signal Processing, Vol. 110, pp. 188-198, 2015. https://doi.org/10.1016/j.sigpro.2014.10.030
  11. X. Wang, T. X. Han, S. Yan, "An HOG-LBP Human Detector with Partial Occlusion Handling," ICCV, pp. 32-39, 2009.
  12. S. Zhang, L. Wen, X. Bian, Z. Lei, S. Z. Li, "Occlusion-Aware R-CNN: Detecting Pedestrians in a Crowd," ECCV, pp. 657-674, 2018.
  13. J. Wang, L. Xie, A.L. Yuille, Z. Zhang, C. Xie, "Deepvoting: A Robust and Explainable Deep Network for Semantic Part Detection Under Partial Occlusion," CVPR, pp. 1372-1380, 2018.
  14. C. Chi, S. Zhang, J. Xing, Z. Lei, S. Z. Li, X. Zou, "PedHunter: Occlusion Robust Pedestrian Detector in Crowded Scenes," AAAI, pp. 10639-10646, 2020.
  15. V. Mnih, N. Heess, A. Graves, K. Kavukcuoglu, "Recurrent Models of Visual Attention," NIPS, pp. 2204-2212, 2014.
  16. Y. Pang, J. Xie, M. H. Khan, R. M. Anwer, F. S. Khan, L. Shao, "Mask-Guided Attention Network for Occluded Pedestrian Detection," ICCV, pp. 4967-4975, 2018.
  17. S. Ren, K. He, R. Girshick, J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," NIPS, pp. 91-99, 2015.
  18. K. He, X. Zhang, S. Ren, J. Sun, "Deep Residual Learning for Image Recognition," CVPR, pp. 770-778, 2016.
  19. P. F. Felzenszwalb, R. B. Girshick, D. A. McAllester, D. Ramanan, "Object Detection with Discriminatively Trained Part-based Models," TPAMI, Vol. 32, No. 9, pp. 1627-1645, 2010. https://doi.org/10.1109/TPAMI.2009.167
  20. R. Girshick, "Fast R-CNN," ICCV, pp. 1440-1448, 2015.
  21. M. Cimpoi, S. Maji, I. Kokkinos, S. Mohamed, A. Vedaldi, "Describing Textures in the Wild," CVPR, pp. 3606-3613, 2016.