Fast intra mode decision algorithm based on texture orientation detection in HEVC

• The texture orientation is detected by computing the local directional variance along a set of co-lines.
• The co-lines use a set of twelve rational slopes with different orientation to the HEVC angular prediction directions.
• The directional variance is computed in a sliding window including the reference pixels.
• According with the dominant gradient detected, a reduced set of candidate modes are selected to be further evaluated.

The High Efficiency Video Coding is the new video coding standard, which obtains over 50% bit rate savings compared with H.264/AVC for the same perceptual quality. Intra-prediction coding in HEVC achieves high coding performance at the expense of high computational complexity, due to the high density of angular prediction modes, among other new features. This work presents a novel texture orientation detection algorithm by computing the local directional variance along a set of co-lines with rational slopes in the discrete lattice Λ. According with the dominant gradient detected, a reduced set of directional candidate modes is selected to be further checked by the rate-distortion optimization stage. The experimental results demonstrate the proposed algorithm can achieve a 30% computational complexity reduction compared with the HEVC reference model, with a negligible bit rate increase of 0.4%, and quality losses lower 0.03 dB. Comparisons with similar state-of-the-art works show the proposed approach achieves the best trade-off between complexity reduction and rate-distortion.

Figure optionsDownload as PowerPoint slide