A Fast Algorithm Of HEVC CU Partition And Inter CU Prediction Mode Decision

High efficiency video coding(High Efficiency Video Coding,HEVC)compared with the previous generation video coding standard H.264 / AVC improves the video compression rate by 50%,and the subjective quality is also equivalent.Its coding framework uses H.264 / AVC technology,which adds more advanced coding technology.Complex coding technology brings a high computational burden,which brings great challenges to the promotion and practical application of technology.In order to reduce the coding time complexity,this paper deeply studies the HEVC coding framework,optimizes the time-consuming CU division process and the inter-frame prediction module,and proposes quick decision method based on the depth range of the HEVC CU and fast inter prediction algorithm based on depth characteristics.HEVC's CU division process is based on the quadtree division method for recursive division.The computational complexity of encoding a CTU is very high.To optimize the CU division process,the neighborhood depth information is used to predict the current CU depth range.HEVC's fast decision method of CU depth range.Compared with the standard test software HM16.9 under Lowdelay configuration,this algorithm can save 32.12% of coding time on average.The inter-frame prediction mode traversal also greatly affects the encoding speed.It needs to traverse 8 different PU prediction modes.To optimize the inter-frame prediction mode,this paper analyzes the inter-frame PU mode of the video sequence in different scenes and different depths.Characteristic fast inter-frame prediction algorithm.Compared with the standard test software HM16.9 under Lowdelay configuration,this algorithm can save an average of 28.6% of coding time.There are two innovations in this paper: 1)In the CU division,this paper calculates the weight by calculating the depth of each CU in the co-located CTU in the time domain,using the number of each depth and the influence of the adjacent CTU in the air domain on the current CTU,and calculating the weight judging the current CTU coding range,avoiding unnecessary calculations based on the prediction depth range based on the maximum and minimum values of the neighborhood and the bit error rate caused by local optimization;2)In inter-frame PU prediction,the first The PU size is divided into regular blocks and irregular blocks,and then the depth information is used to determine the complexity of the current CTU,and then the CU blocks of different depths are partitioned,and the partitions with small depths are traversed by the PU regular blocks,and the partitions with large depths are based on The complexity is refined to ensure a small bit error rate.The final experimental results show that the coding efficiency can be greatly improved.This paper proposes that the above two algorithms can greatly save coding time,and the bit loss is negligible,which is of great help to the improvement of coding time efficiency.The algorithm can provide some reference for real-time transmission and storage of high-resolution video in the next 4K,8K,improve transmission efficiency and save storage space.