Efficient Partition Decision Based on Visual Perception and Machine Learning for H.266/Versatile Video Coding

H.266/Versatile Video Coding (VVC) is the latest international video coding standard to encode ultra-high-definition effectively. The quadtree with nested multi-type tree (QT-MTT) structure provides various sizes of partitioning and allows binary (BT) split ternary (TT) at each QT level. Furthermore, numerous advanced tools are equipped in H.266/VVC encoder. However, encoding time increases tremendously. Previous researches regarding fast algorithm seldom mention perceptual redundancy. This paper utilizes human vision model just noticeable difference extract visually distinguishable pixels that may affect visual perception. We observe distributions acquired by horizontal vertical projections within unit related their corresponding MTT splitting modes. Therefore, representing information used be input features machine learning. Fast decision determined random forest models learning proposed quickly select partition for intra coding. Experimental results demonstrate method can effectively accelerate process while maintaining good bitrate quality based on properties better performance than previous work.