Abstract
Selection of motion vector (MV) has a significant impact on the quality of an encoded, and particularly a transcoded video, in terms of rate-distortion (R-D) performance. The conventional motion estimation process, in most existing video encoders, ignores the rate of residuals by utilizing rate and distortion of motion compensation step. This approach implies that the selected MV depends on the quantization parameter. Hence, the same MV that has been selected for high bit rate compression may not be suitable for low bit rate ones when transcoding the video with motion information reuse technique, resulting in R-D performance degradation. In this paper, we propose an R-D optimized motion selection criterion that takes into account the effect of residual rate in MV selection process. Based on the proposed criterion, a new two-piece Lagrange multiplier selection is introduced for motion estimation process. Analytical evaluations indicate that our proposed scheme results in MVs that are less sensitive to changes in bit rate or quantization parameter. As a result, MVs in the encoded bitstream may be used even after the encoded sequence has been transcoded to a lower bit rate one using re-quantization. Simulation results indicate that the proposed technique improves the quality performance of coding and transcoding without any computational overhead.
Highlights
In ubiquitous multimedia access, the same content is transferred between different users with various devices and applications; each has different capabilities in their resources such as network bandwidth, display resolution, frame rate, and decoding bit rate
We show that our proposed scheme results in motion vectors that are less sensitive to changes in bit rate or QP
Where rate of the candidate motion vectors (RMC) and DMC are the number of bits required to code the motion vector and the distortion of motion compensation, respectively, λ is the Lagrange multiplier which is determined based on QP, and SA is the search area
Summary
The same content is transferred between different users with various devices and applications; each has different capabilities in their resources such as network bandwidth, display resolution, frame rate, and decoding bit rate. The more recent motion estimation algorithms, such as the one adopted in the joint model (JM) of the H.264/ AVC video encoder, are based on the methods proposed in [6] and [13] In this method, MV is selected by minimizing the motion compensation cost, as defined in (2), MV 1⁄4 arg min fJMC 1⁄4 DMCðmvÞ þ λRMCðmvÞg mv∈SA ð2Þ where RMC and DMC are the number of bits required to code the motion vector and the distortion of motion compensation, respectively, λ is the Lagrange multiplier which is determined based on QP, and SA is the search area. This model separates the effects of motion compensation and residual coding processes on the distortion of the block
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
