Journal of Broadcast Engineering (방송공학회논문지)

The Korean Institute of Broadcast and Media Engineers (한국방송∙미디어공학회)
권호 표지
DOI QR코드

DOI QR Code

Activated Viewport based Surveillance Event Detection in 360-degree Video

360도 영상 공간에서 활성 뷰포트 기반 이벤트 검출

  • Shim, Yoo-jeong (Dept. of Electronics and Information Engineering, Korea Aerospace University) ;
  • Lee, Myeong-jin (Dept. of Electronics and Information Engineering, Korea Aerospace University)
  • 심유정 (한국항공대학교 항공전자정보공학과) ;
  • 이명진 (한국항공대학교 항공전자정보공학과)
  • Received : 2020.07.01
  • Accepted : 2020.08.28
  • Published : 2020.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Since 360-degree ERP frame structure has location-dependent distortion, existing video surveillance algorithms cannot be applied to 360-degree video. In this paper, an activated viewport based event detection method is proposed for 360-degree video. After extracting activated viewports enclosing object candidates, objects are finally detected in the viewports. These objects are tracked in 360-degree video space for region-based event detection. The proposed method is shown to improve the recall and the false negative rate more than 30% compared to the conventional method without activated viewports.

360도 영상은 일반 화각 영상과 다른 프레임 구조를 갖기 때문에 기존 영상 보안 이벤트 검출 알고리즘들을 360도 영상에 바로 적용 시 왜곡에 의한 성능 저하가 발생할 수 있다. 본 논문에서는 360도 영상 공간에서 활성 뷰포트 기반 이벤트 검출 기법을 제안한다. 다중 객체 영역들을 포함하는 활성 뷰포트들을 생성하고, 뷰포트 내 객체 검출과 360도 영상 공간에서 객체 추적을 수행하며, 영역 기반의 보안 이벤트를 검출한다. 제안 방법은 360도 ERP 시퀀스들에 대해 성능이 평가되었고, 활성 뷰포트를 사용하지 않은 이벤트 검출 방법보다 30% 이상의 recall, 30% 이상의 false negative rate 성능 향상을 보였다.

Keywords

References

  1. J. Redmon and A. Farhadi, "Yolov3: An incremental improvement," arXiv preprint arXiv:1804.02767, 2018.
  2. N. Wojke, A. Bewley, and D. Paulus, "Simple online and realtime tracking with a deep association metric," IEEE International Conference on Image Processing, pp. 3464-3468, Sep 2017.
  3. Lee, Yeonkun, Jaeseok Jeong, Jongseob Yun, Wonjune Cho, and Kuk-Jin Yoon, "SpherePHD: Applying CNNs on a Spherical PolyHeDron Representation of $360^{\circ}$ Images," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9181-9189, 2019.
  4. Su, Yu-Chuan, and Kristen Grauman. "Kernel transformer networks for compact spherical convolution," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9442-9451, 2019.
  5. Yoo-jeong Shim, Min-seok Lee, Seok-ho Baek , Myeong-jin Lee. "Multi-object tracking method and system for 360-degree video," Summer Annual conference of IEIE, Jun:700-701, 2019.
  6. Nayak, Rashmiranjan, Mohini Mohan Behera, Vaisyaraju Girish, Umesh Chandra Pati, and Santos Kumar Das, "Deep Learning Based Loitering Detection System Using Multi-Camera Video Surveillance Network," IEEE International Symposium on Smart Electronic Systems, pp. 215-220, 2019.
  7. Park, Keon-woo, Doo-sik Kang, Jun-sik Kim, and Myeong-jin Lee, "Loitering Detection based on Regional Histogram and Object Velocity," Summer Annual conference of KICS, pp.1304-1305, 2016.
  8. Kim, Seajin, Byung-jin Lee, Jae-won Jeong, and Myeong-jin Lee. "Multi-object tracking coprocessor for multi-channel embedded DVR systems," IEEE Transactions on Consumer Electronics, pp.1366-1374, 2012. https://doi.org/10.1109/TCE.2012.6415008
  9. Korea Internet&Security Agency, "Intelligent CCTV Solution Performance Test and Certification Guide," pp. 30, 2018.
  10. Bernardin, Keni, and Rainer Stiefelhagen. "Evaluating multiple object tracking performance: the CLEAR MOT metrics," EURASIP Journal on Image and Video Processing, pp. 1-10, Dec 2008.