Complexity Analysis Of Next-Generation VVC Encoding And Decoding

Abstract

While the next generation video compression standard, Versatile Video Coding (VVC), provides a superior compression efficiency, its computational complexity dramatically increases. This paper thoroughly analyzes this complexity for both encoder and decoder of VVC Test Model 6, by quantifying the complexity break-down for each coding tool and measuring the complexity and memory requirements for VVC encoding/decoding. These extensive analyses are performed for six video sequences of 720p, 1080p, and 2160p, under Low-Delay (LD), Random-Access (RA), and All-Intra (AI) conditions (a total of 320 encoding/decoding). Results indicate that the VVC encoder and decoder are 5x and 1.5x more complex compared to HEVC in LD, and 31x and 1.8x in AI, respectively. Detailed analysis of coding tools reveals that in LD on average, motion estimation tools with 53%, transformation and quantization with 22%, and entropy coding with 7% dominate the encoding complexity. In decoding, loop filters with 30%, motion compensation with 20%, and entropy decoding with 16%, are the most complex modules. Moreover, the required memory bandwidth for VVC encoding/decoding are measured through memory profiling, which are 30x and 3x of HEVC. The reported results and insights are a guide for future research and implementations of energy-efficient VVC encoder/decoder.

Comparative Rate-Distortion-Complexity Analysis of VVC and HEVC Video Codecs

“…Table 1 highlights the key differences between the existing comparisons and ours. Related works can be classified into three categories: 1) RD comparisons [19], [20]; 2) complexity comparisons [21], [22]; and 3) RDC comparisons [23]- [26].…”

“…In [22], VTM6.0 was used to quantify the average shares of different encoding and decoding tool categories. HM served as an anchor.…”

Complexity Analysis of a Versatile Video Coding Decoder over Embedded Systems and General Purpose Processors

“…Decoding ratios of 1.7 to 4.7 fps and of 2.1 to 8.4 fps were obtained for unoptimized and optimized decoder versions, respectively. In [ 5 ], a core i7-4790K 4.0 GHz HGPP was used to profile an implementation of a VTM6.0 decoder, but the global fps or processing times were not reported in this paper. As a result, it would not have been possible to make a fair comparison in terms of fps, but later in this section, it is used to compare the computational load among functional blocks.…”

“…Similar to HEVC, the merge mode adopts candidates from spatial, temporal, and zero motion vectors. Further, it adopts a pairwise average vector, and a history-based vector [ 5 ]. VVC has achieved better precision accuracy than its predecessors by performing motion prediction at the sub-coding unit level.…”

“…Various new features have been adapted to VVC to achieve better coding efficiency [ 5 ]. For example, the largest size of coding tree units (CTUs) is 128 × 128 pixels, quadtree with nested multi-type tree (QTMT) for coding unit (CU) partition, multiple transform selection (MTS) support up to 87 intra coding modes [ 6 ], cross-component linear model, affine motion compensation [ 7 ] and adaptive loop filter (ALF).…”