“…For comparison, Table 3 lists the key performance of state-of-the-art ASIC and FPGA-based works, including the VVC-MTS related works [11,9,6,10]. Gate count is the logical calculation part and it can be seen from Table 3 that compared with implementations of Fan et al [11] and Mert et al [6], our solution has obvious advantages. We present a unified transform architecture that can realize IDCT-II/IDST-VII/IDCT-VIII for transform unit of order 4,8,16,32 and 64 with a fixed throughput of 2 pixels per cycle.…”