Proceedings of the Korean Society of Computer Information Conference (한국컴퓨터정보학회:학술대회논문집)

Korean Society of Computer Information (한국컴퓨터정보학회)
권호 표지

Unveiling the Unseen: A Review on current trends in Open-World Object Detection

오픈 월드 객체 감지의 현재 트렌드에 대한 리뷰

  • MUHAMMAD ALI IQBAL (Department of Computer Engineering, Jeju National University) ;
  • Soo Kyun Kim (Department of Computer Engineering, Jeju National University)
  • Published : 2024.01.17
Download PDF
⟨ Previous Next ⟩

Abstract

This paper presents a new open-world object detection method emphasizing uncertainty representation in machine learning models. The focus is on adapting to real-world uncertainties, incrementally updating the model's knowledge repository for dynamic scenarios. Applications like autonomous vehicles benefit from improved multi-class classification accuracy. The paper reviews challenges in existing methodologies, stressing the need for universal detectors capable of handling unknown classes. Future directions propose collaboration, integration of language models, to improve the adaptability and applicability of open-world object detection.

Keywords

Acknowledgement

이 논문은 2021년도 정부(교육부)의 재원으로 한국연구재단의 지원을 받아 수행된 기초연구사업임(No. 2021R1I1A3058103)

References

  1. W. J. Scheirer, A. de Rezende Rocha, A. Sapkota, and T. E. Boult, "Toward open set recognition," IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 7, pp. 1757-1772, Jul. 2013.  https://doi.org/10.1109/TPAMI.2012.256
  2. L. P. Jain, W. J. Scheirer, and T. E. Boult, "Multi-class open set recognition using probability of inclusion," in Computer Vision-ECCV 2014. Cham, Switzerland: Springer, 2014, pp. 393-409. 
  3. S. D. Z. Ge and R. Garnavi, "Generative OpenMax for multi-class open set classification," in Proc. Brit. Mach. Vis. Conf., Sep. 2017, pp. 42.1-42.12. 
  4. A. R. Dhamija, M. Gunther, J. Ventura, and T. E. Boult, "The overlooked elephant of object detection: Open set," in Proc. IEEE Winter Conf. Appl. Comput. Vis. (WACV), Mar. 2020, pp. 1010-1019 
  5. O. Zohar, K.-C. Wang, and S. Yeung, "PROB: Probabilistic objectness for open world object detection," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2023, pp. 11444-11453. 
  6. Z. Wu, Y. Lu, X. Chen, Z. Wu, L. Kang, and J. Yu, "UC-OWOD: Unknown-classified open world object detection," in Computer Vision-ECCV 2022, S. Avidan, G. Brostow, M. Cisse, G. M. Farinell and T. Hassner, Eds. Cham, Switzerland: Springer, 2022, pp. 193-210. 
  7. Y. Ma, H. Li, Z. Zhang, J. Guo, S. Zhang, R. Gong, and X. Liu, "Annealingbased label-transfer learning for open world object detection," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2023, pp. 11454-11463 
  8. K. J. Joseph, S. Khan, F. S. Khan, and V. N. Balasubramanian, "Towards open world object detection," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2021, pp. 582