Performance of multi camera views' detection using MPEG-7 visual signature tools

• Autorzy: Wnukowicz, K. , Galiński, G.
• Języki publikacji: EN

Abstrakty

EN

The paper presents experimental results for the detection of related video streams (views) of multi camera system using MPEG-7 visual signature tools. This detection can be used for identification of video streams stored in video repositories which constitute the same multi camera recording set. Two signature tools have been investigated: Image Signature descriptor and Video Signature descriptor. The motivation for the usage of the tools designed primarily for replica detection is the fact that the video material produced by multi camera system typically consists of a few related video streams which are similar to each other and, thus, the video streams produced by multi camera system can be regarded as semi-replicas as they contain predominantly the same scenes or objects recorded at the same time. Experiments show that only in some restricted scenarios of a multi camera system setup the MPEG-7 visual signature tools can be used for reliable detection of multi camera views. These reliable scenarios include systems having cameras with perpendicular optical axes and "far" scene.

Słowa kluczowe

EN

multi view video detection; multi camera systems; MPEG-7 visual signature tools

• Czasopismo: Opto-Electronics Review
• Rocznik: 2013
• Tom: Vol. 21, No. 1
• Strony: 79-85
• Opis fizyczny: Bibliogr. 9 poz., rys., il.

Twórcy

autor

Wnukowicz, K.

• Institute of Radioelectronics, Warsaw University of Technology, 15/19 Nowowiejska Str., 00-665 Warszawa, Poland,

autor

Galiński, G.

• Institute of Radioelectronics, Warsaw University of Technology, 15/19 Nowowiejska Str., 00-665 Warszawa, Poland

Bibliografia

• 1. B. S. Manjunath, P. Salembier, and T. Sikora,Introduction to MPEG-7: Multimedia Content Description Interface, edited by John Wiley & Sons Inc., New York, 2002.
• 2. M. Bober and P. Brasnett, „MPEG-7 visual signature tools”, IEEE I. Conf. Multimedia and Expo, pp. 1540–1543, New York, 2009.
• 3. P. Brasnett, S. Paschalakis, and M. Bober, „Recent developments on standardisation of MPEG-7 visual signature tools”, IEEE I. Conf. Multimedia and Expo, pp. 1347–1352, Singapore, 2010.
• 4. P. Brasnett and M. Bober, „Fast and robust image identification”, 19th Int. C. Patt. Recog., pp. 871–876, Tampa, 2008.
• 5. P. Brasnett and M. Bober, „A Robust visual identifier using the trace transform”, Proc. Visual Information Engineering Conference, pp. 25–27, London, 2007.
• 6. M. Bober and S. Paschalakis, „MPEG image and video signature”, in MPEG Representation of Digital Media, pp. 81–95, edited by L. Chiariglione, Springer, New York, 2012.
• 7. A. Vetro, M. McGuire, W. Matusik, A. Behrens, J. Lee, and H. Pfister, „Multiview video test sequences from MERL”, ISO/IEC JTC1/SC29/WG11 doc. M12077, Busan, Korea, 2005.
• 8. M. Domański, T. Grajek, K. Klimaszewski, M. Kurc, O. Stankiewicz, J. Stankowski, and K. Wegner, „Poznań multiview video test sequences and camera parameters”,ISO/IEC JTC1/SC29/WG11 doc. M17050, Xian, China, 2009.
• 9. A. Smolic, „MPEG-3DAV FhG-HHI Test Data”, available at https://www.3dtv-research.org/downloadlist.php-dat&s=61