Fast motion vector recovery algorithm in H.264 video streams

Kavish Seth; V. Kamakoti; S. Srinivasan

doi:10.1117/12.843973

27 January 2010 Fast motion vector recovery algorithm in H.264 video streams

Kavish Seth, V. Kamakoti, S. Srinivasan

Proceedings Volume 7542, Multimedia on Mobile Devices 2010; 75420N (2010) https://doi.org/10.1117/12.843973
Event: IS&T/SPIE Electronic Imaging, 2010, San Jose, California, United States

Abstract

This paper proposes a fast statistical approach to recover lost motion vectors in H.264 video coding standard. Unlike other video coding standards, the motion vectors of H.264 cover smaller area of the video frame being encoded. This leads to a strong correlation between neighboring motion vectors, thus making H.264 standard amenable for statistical analysis to recover the lost motion vectors. This paper proposes a Pearson Correlation Coefficient based matching algorithm that speeds up the recovery of lost motion vectors with very less compromise in visual quality of the recovered video. To the best of our knowledge, this is the first attempt that employs correlation coefficient for motion vector recovery. Experimental results obtained by employing the proposed algorithm on standard benchmark video sequences show that they yield comparable quality of recovered video with significantly less computation than the best reported in the literature, thus making it suitable for real-time applications.

Citation Download Citation

Kavish Seth, V. Kamakoti, and S. Srinivasan "Fast motion vector recovery algorithm in H.264 video streams", Proc. SPIE 7542, Multimedia on Mobile Devices 2010, 75420N (27 January 2010); https://doi.org/10.1117/12.843973

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available