Pipelined hardware implementation of sub-pixel motion estimation reduces LUT usage in HEVC, indicating enhanced efficiency.
High Efficiency Video Coding (HEVC) is the most widely used Video encoder which outperforms prior standards. In HEVC, motion estimation unit is used to exploit the temporal redundancy between successive frames. It consists of integer and sub-pixel motion estimation. This unit consumes massive hardware resources due to the intensive computations there. In this paper, we present an efficient hardware implementation of sub-pixel motion estimation unit for HEVC. It is a pipelined structure with a limited number of computations to mitigate the hardware demand for HEVC encoders. The proposed architecture was prototyped in Verilog HDL and synthesized using Altera Cyclone V FPGA. The reports show that the proposed design achieves large saving in hardware resources compared with existing approaches while preserving quality and real time consideration. It takes 1,165 LUT and 1,103 registers from Cyclone V FPGA. The proposed design can achieve a maximum frequency of 258 MHz to cope with the real-time applications. It can process 107, 60, 27 fps for 1K, 2k, and 4k resolution, respectively. It outperforms other schemes in literature with hardware reduction up to 94.15% in the number of LUTs.
No takes yet. Share an insight, caveat, or question.
Abdelaziz et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: