Leveraging cuboids for better motion modeling in high efficiency video coding

Ashek Ahmmed, Manzur Murshed, Manoranjan Paul

Research output: Book chapter/Published conference paperConference paper

Abstract

In conventional video compression systems, motion model is used to approximate the geometry of moving object boundaries. It is possible to relieve motion model from describing discontinuities in the underlying motion field, by incorporating motion hint that can predict the spatial structure of future frames using the structure of reference frames. However, formation of highly accurate motion hint is computationally demanding, in particular for high resolution video sequences. Cuboids, rectangular regions derived using statistical features, attempt to separate out different objects present in the scene; they are computationally efficient and have sparse representation. Leveraging on the advantages of cuboids, in this paper, we propose to discover homogeneous motion regions and their associated motion based on cuboids. Afterwards, the estimated motion models and their domains are applied to form a prediction of the current frame. Experimental results show that a savings in bit rate of 3.96% is achievable over standalone HEVC reference, if this predicted frame is used as an additional reference frame for the current frame.
Original languageEnglish
Title of host publication2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages2188-2192
Number of pages5
ISBN (Electronic)9781509066315
ISBN (Print)9781509066322
DOIs
Publication statusPublished - May 2020
Event2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - virtual conference, Barcelona, Spain
Duration: 04 May 202008 May 2020
https://2020.ieeeicassp.org/ (conference website)
https://cmsworkshops.com/ICASSP2020/TechnicalProgram.asp (program)
https://2020.ieeeicassp.org/authors/paper-kit/ (peer review process)

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2020-May
ISSN (Print)1520-6149

Conference

Conference2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020
CountrySpain
CityBarcelona
Period04/05/2008/05/20
OtherICASSP is the world’s largest and most comprehensive technical conference focused on signal processing and its applications. The 2020 conference will feature world-class presentations by internationally renowned speakers, cutting-edge session topics and provide a fantastic opportunity to network with like-minded professionals from around the world.

ICASSP 2020 is going to have over 3000 participants which make this year’s event the largest ever ICASSP conference over the past 45 years. This will provide a great opportunity and arena for industry participation and exhibition. During this conference, you will visit the world largest international companies and publishers who will also sponsor a number of prestigious awards. In addition, there is a significant record in tutorial registrations as over 1200 people have registered for the tutorial sessions to be held in the first two days of the conference.
Internet address

Fingerprint Dive into the research topics of 'Leveraging cuboids for better motion modeling in high efficiency video coding'. Together they form a unique fingerprint.

  • Cite this

    Ahmmed, A., Murshed, M., & Paul, M. (2020). Leveraging cuboids for better motion modeling in high efficiency video coding. In 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2020 - Proceedings (pp. 2188-2192). [9053851] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2020-May). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICASSP40776.2020.9053851