FPGA implementations of HEVC Inverse DCT using high-level synthesis

Abstract

High Efficiency Video Coding (HEVC), the recently developed international video compression standard, has 50% better video compression efficiency than H.264 video compression standard at the expense of significantly increased computational complexity. HEVC Inverse Discrete Cosine Transform (IDCT) algorithm accounts for 11% of the computational complexity of an HEVC video encoder. Recently, commercial and academic high-level synthesis (HLS) tools are started to be successfully used for FPGA implementations of digital signal processing algorithms. Therefore, in this paper, the first FPGA implementations of HEVC 2D IDCT algorithm using HLS tools in the literature are proposed. The proposed HEVC IDCT hardware are implemented on Xilinx FPGAs using three HLS tools; Xilinx Vivado HLS, LegUp, MATLAB Simulink HDL Coder. Using HLS tools significantly reduced the FPGA development time, and the resulting FPGA implementations achieved real-time performance. Therefore, HLS tools can be used for FPGA implementation of HEVC video encoder.