Optimization of High Efficiency Video Coding Integer Motion Estimation Algorithm Based on Hardware Implementation
Keywords:
High-efficiency video coding, Integer-pixel motion estimation, Video coding standard, Hardware implementation, Small diamond search algorithmAbstract
In order to reduce the number of cycles of the motion estimation module and improve hardware efficiency, optimizes the small diamond search algorithm from the perspective of hardware implementation. First, the iteration and processing sequence of PU and CU were adjusted to solve the problem of pipeline stagnation in the process of processing. At the same time, the parallel computing method for small PU blocks can further improve the processing speed. This article first uses MATLAB to implement the search algorithm, uses Verilog language to describe the hardware circuit, the two versions use the same excitation file on the data, and compare the intermediate values of each module for functional verification. Through the test of multiple sets of sequences, the hardware circuit of this article needs an average consumption of 5,800 clk for AVMP and IME processing on a 64 px ´ 64 px CTU. The Arria10AX115N3F40E2SG development board is selected on the QuartusII platform, the main frequency can reach 186 MHz, and the overall performance of the whole pixel motion estimation module can reach 1080p @ 61 fs-1.
References
Sullivan GJ, Ohm JR, Han WJ, et al., 2013, Overview of the High Efficiency Video Coding (HEVC) Standard. IEEE Transactions on Circuits and Systems for Video Technology, 22(12): 1649–1668.
Wan S, Yang F, 2014, New Generation High-Efficiency Video Coding H.265/HEVC: Principles, Standards, and Implementation. Publishing House of Electronics Industry, Beijing.
Song B, Chang Y, Zhou N, 2006, A Fast Algorithm Based on H.264 Inter-frame Prediction. Acta Electronica Sinica, 34(1): 31–34.
Jain JR, Jain AK, 1981, Displacement Measurement and Its Application in Interframe Image Coding. IEEE Transactions on Communications, 29(12): 1799–1808.
Koga T, 1981, Motion Compensated Interframe Coding for Video Conferencing. IEEE Proceedings of the National Telecommunication Conference, New Orleans, 1981: 531–535.
Po LM, Ma WC, 1996, A Novel Four-Step Search Algorithm for Fast Block Motion Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 6(3): 313–317.
Shen LQ, Liu Z, Zhang XP, et al., 2013, An Effective CU Size Decision Method for HEVC Encoders. IEEE Transactions on Multimedia, 15(2): 465–470.
Fan YB, Huang LL, Hao B, et al., 2018, A Hardware-Oriented IME Algorithm for HEVC and Its Hardware Implementation. IEEE Transactions on Circuits and Systems for Video Technology, 28(8): 2048–2057.
Estefania A, Roberto G, Otoniel ML G, et al., 2019, Design and Implementation of an Efficient Hardware Integer Motion Estimator for an HEVC Video Encoder. Journal of Real-Time Image Processing, 16(2): 547–557.
Cheng X, Liu ZY, Tetsunori K, et al., 2016, Multi-Feature Based Fast Depth Decision in HEVC Inter Prediction for VLSI Implementation. The 9th International Congress on Image and Signal Processing, Biomedical Engineering, and Informatics, Datong, IEEE, 2016: 124–128.
Li GL, Wang CC, Chiang KH, 2014, An Efficient Motion Vector Prediction Method for Avoiding AMVP Data Dependency for HEVC. IEEE International Conference on Acoustics, Speech, and Signal Processing, Florence, IEEE, 2014: 7363–7366.
Chen Q, Shi L, Huang B, et al., 2018, A Fast Algorithm and Hardware Architecture for Motion Estimation in HEVC. Journal of Fuzhou University (Natural Science Edition), 46(5): 636–643.
Gao X, 2018, Research on Fast Algorithm and Hardware Implementation of Inter-Prediction Mode Selection in HEVC, thesis, Fuzhou University.
Ye X, Ding DD, Yu L, 2014, A Hardware-Oriented IME Algorithm and Its Implementation for HEVC. IEEE Visual Communications and Image Processing Conference, Valletta, IEEE, 2014: 205–208.
Zhang Y, Li Q, 2018, A Fast Decision Algorithm for HEVC Inter-frame Mode Based on Motion Characteristics. Computer Engineering and Applications, 54(23): 195–202.
Gogoi S, Peesapati R, 2021, Design and Implementation of an Efficient Multi-Pattern Motion Estimation Search Algorithm for HEVC/H.265. IEEE Transactions on Consumer Electronics, 67(4): 319–328.