All-intra Configuration Research Articles

Fast HEVC to SCC Transcoder by Early CU Partitioning Termination and Decision Tree-Based Flexible Mode Decision for Intra-Frame Coding

Screen content coding (SCC) was developed to enhance High Efficiency Video Coding (HEVC) for encoding screen content videos. However, HEVC has dominated the market for many years, and it leaves many legacy screen content videos encoded by HEVC. Therefore, it is desired that the legacy screen content videos are migrated from HEVC to SCC to improve the coding efficiency. This paper presents a fast transcoding algorithm by analyzing various features from four categories. They are the features from the HEVC decoder, static features, dynamic features, and spatial features. First, the coding unit (CU) depth level collected from the HEVC decoder is utilized to early terminate the CU partition in SCC. Second, a flexible encoding structure is proposed to make early mode decisions with the help of various features. On the one hand, high decision accuracy is achieved because mode decision is considered from different aspects by utilizing features from more than one category. On the other hand, high computational complexity is reduced because the flexible structure considers the decision of each mode separately. The experimental results show that the proposed algorithm provides 51.24% and 54.65% re-encoding time reduction with 1.32% and 1.25% negligible Bjontegaard delta bitrate loss for YUV 4:2:0 and YUV 4:4:4 screen content sequences using all-intra configuration, respectively.

Efficient Intra Bitrate Transcoding for Screen Content Coding Based on Convolutional Neural Network

The Screen Content Coding (SCC) extension of High Efficiency Video Coding (HEVC) is developed to improve the coding efficiency of screen content videos. To meet the diverse network requirement of different clients, bitrate transcoding for SCC is desired. This problem can be solved by a conventional brute-force transcoder (CBFT) which concatenates an original decoder and an original encoder. However, it induces high computational complexity associated with the re-encoding part of CBFT. This paper presents a convolutional neural network based bitrate transcoder (CNN-BRT) for SCC. By utilizing information from both the decoder side and the encoder side, CNN-BRT makes a fast prediction for all coding units (CUs) of a coding tree unit (CTU) in a single test. At the decoder side, decoded optimal mode maps that reflect the optimal modes and CU partitions in a CTU is derived. At the encoder side, the raw samples in a CTU are collected. Then, they are fed to CNN-BRT to make a fast prediction. To imitate the optimal mode selection in the original re-encoding part, CNN-BRT involves a loss function that takes both of the sub-optimal modes and the final optimal modes into consideration. Compared with the HEVC-SCC reference software SCM-3.0, the proposed CNN-BRT reduces encoding time by 54.86% on average with a negligible Bjontegaard delta bitrate increase of 1.01% under all-intra configuration.

A computationally scalable fast intra coding scheme for HEVC video encoder

The High Efficiency Video Coding (HEVC) adopts 35 intra prediction modes to provide a more precise intra prediction. As a result, HEVC intra coding is much more complex. To provide a flexible solution for battery-powered devices, which include varying power budgets based on battery level, this paper jointly employs statistical distribution of different intra modes, and texture analysis. While the statistical information specifies the most efficient modes for extremely low power budgets, for moderate to high power budgets the best angular modes are predicted by analyzing the residual of the Planar mode, which is demonstrated to represent the dominant edges of the block. Moreover, a scheme is presented that further reduces the complexity by predicting the more important DC and Planar modes using the information provided by the entropy coder. The proposed method imposes almost no computational overhead since it utilizes the existing internal operators of HEVC. Ultimately, an intra prediction scheme with seven predefined levels of complexity is presented that provides coding for various power budgets and devices with various capabilities. The highest level of this scheme provides ~23% total encoding time reduction compared to HM, in all-intra configuration, with only 0.78% penalty on BD-Rate for high end devices. While the lowest level makes encoding possible for extremely low power devices and low battery plans with ~52% time reduction (almost the entire share of intra coding) at the cost of 9.6% BD-Rate, which is shown to be nearly minimal for this level of processing power.

A fast CU partition and mode decision algorithm for HEVC intra coding

The latest video coding standard H.265/HEVC is a milestone because it can achieve up to 50 % coding performance improvement compared to the widely used H.264/AVC, owing to the newly introduced coding tools. However, these new techniques also introduce dramatic computational complexity increase. In order to release the computational burden while maintaining the coding efficiency, a two-part fast intra coding algorithm is proposed in this paper. First a CTU depth range prediction and two Bayesian decision rules are applied to accelerate the CU partition process. The depth search range of the current CTU is restricted by the neighboring CTUs on account of the spatial correlation. And one early split and one early termination strategies is presented based on the minimum Bayesian decision rule. Next, a three-step fast intra mode decision algorithm is performed to speed up the intra coding, both the depth information and the orientation information of the current CU are used to skip some unnecessary intra mode check, and the order of Most Probable Modes (MPM) is adjusted to further remove redundant calculation. As demonstrated in the experimental results, our proposed algorithm can reduce 52.9% encoding time on average with the cost of an negligible coding efficiency loss under the All-Intra configuration.

Fast mode decision algorithm for HEVC intra coding based on texture partition and direction

High efficiency video coding (HEVC) is the newest video coding standard, which employs some advanced coding techniques as compared to the previous standard H.264. The flexible quad-tree partitioning of coding tree unit (CTU) and various candidate modes of prediction unit (PU) significantly promote the video compression efficiency; however, these techniques lead to a great amount of computational loads. In this paper, a fast mode decision algorithm for HEVC intra coding is proposed based on texture partition and direction. It consists of two sub-algorithms: the CTU depth range prediction (CDRP) and the intra-prediction mode selection (IPMS). The CDRP reduces the recursive partition number of coding unit (CU) based on the correlation between the CTU texture partition and the optimum CU partition, and it first calculates the texture partition flags of different-size CUs from bottom to top. Then, it employs these partition flags to predict the depth range of the current CTU and decide whether to terminate the CU partition in advance. In order to reduce the number of candidate PU modes for the Hadamard optimization, the IPMS first uses the three-step selection of the candidate modes. The first step selects the candidate modes based on the correlation between the texture directions and the optimum PU modes. The second step selects the candidate modes by using the best modes among the selected modes in the first step. The third step selects the candidate modes by using the spatial correlation of the optimum modes between the current PU and its adjacent PUs. Then, in order to reduce the number of candidate modes for the rate-distortion optimization, the IPMS utilizes the numerical relationship of the sorted Hadamard costs of above selected modes, the optimum modes of adjacent PUs and the statistical characteristics of the small-size PUs. Compared to the original algorithm in HEVC test model, the proposed overall algorithm can reduce 60% encoding time on average with only a 1.45% increase in Bjontegaard delta bit rate under the all-intra configuration. Compared to the most of state-of-the-art algorithms, the proposed overall algorithm has better computational performances and similar rate-distortion performances.

Fast Intra-Mode and CU Size Decision for HEVC

The latest video coding standard High Efficiency Video Coding (HEVC) achieves about a 50% bit-rate reduction compared with H.264/AVC under the same perceptual video quality. For intra coding, a coding unit (CU) is recursively divided into a quadtree-based structure from the largest CU 64 × 64 to the smallest CU 8 × 8. Also, up to 35 intra-prediction modes are allowed. These two techniques improve the intra-coding performance significantly. However, the encoding complexity increases several times compared with H.264/AVC intra coding. In this paper, fast intra-mode decision and CU size decision are proposed to reduce the complexity of HEVC intra coding while maintaining the rate-distortion (RD) performance. For fast intra-mode decision, a gradient-based method is proposed to reduce the candidate modes for rough mode decision and RD optimization. For fast CU size decision, the homogenous CUs are early terminated first. Then two linear support vector machines that employ the depth difference and HAD cost ratio (and RD cost ratio) as features are proposed to perform the decisions of early CU split and early CU termination for the rest of the CUs. Experimental results show that the proposed fast intra-coding algorithm achieves about a 54% encoding time reduction on average with only a 0.7% BD-rate increase for the HEVC reference software HM 14.0 under all-intra configuration.

Fast Mode and Partition Decision Using Machine Learning for Intra-Frame Coding in HEVC Screen Content Coding Extension

This paper presents a fast mode and partition decision framework for screen content coding (SCC) based on machine learning. Extensive statistical studies and complexity evaluations are conducted to explore the distribution of different coding modes and their complexities. The proposed encoder scheme is designed based on the results from these studies. Firstly, a CU is classified as either a natural image block (NIB) or a screen content block (SCB). An SCB only goes through SCC modes at the current CU level. An NIB is further classified as “ partitioned” or “ non-partitioned”. The partitioned block will bypass current level intra modes. The non-partitioned block is classified as “ directional” or “ non-directional” and only goes through a subset of intra candidates. Decision tree classifiers are designed with chosen features that can distinguish different types of blocks. Furthermore, additional mode/partition checking is terminated once the current mode coding rate is lower than a statistics-based threshold. Compared with HEVC-SCC reference software, our proposed fast encoder can balance the encoding efficiency and complexity by adjusting decision confidence thresholds and rate thresholds. Under all-intra configurations, our “ rate-distortion preserving” setting can achieve 40% complexity reduction with only 1.46% BD-loss. Our “ complexity-reduction boosting” setting can achieve 52% complexity reduction with 3.65% BD-loss.

An AR based fast mode decision for H.265/HEVC intra coding

As a critical technique to raise coding efficiency in H.265/HEVC (H.265/High Efficiency Video Coding), the number of prediction modes for intra coding is much larger than that in H.264/AVC. However, the large number of prediction modes also leads to high complexity. In this paper, an AR (Auto Regression) based fast mode decision method for H.265/HEVC intra coding is proposed, including an AR based fast mode selection and an MPM (Most Probable Mode) joint fast mode selection method. Considering the correlation between SATD (Sum of Absolute Transformed Differences) cost and RD (Rate-Distortion) cost, we establish a selective function controlled by an adaptive parameter which is acquired by an AR model. According to the distribution of the SATD cost, the selective function can decide which modes can be discarded and which modes can be used for the FMD (Fine Mode Decision). In addition, in order to further reduce the number of candidate modes for FMD, an MPM joint fast mode selection method is presented based on the relationship between MPMs and candidate modes. Experimental results show that just in PU layer, the proposed AR based fast intra mode decision method can save encoding time by 32 % with negligible coding loss in all-intra main configuration compared to the intra mode decision of HM13.0.

Fast and efficient intra mode decision for HEVC, based on dual-tree complex wavelet

The new video coding standard, High Efficiency Video Coding (HEVC) uses up to 33 angular intra modes to achieve better coding efficiency in intra coding. While larger number of modes results in a significant increase in coding efficiency than its predecessor H.264/AVC, it leads to much more complexity. To reduce this complexity, this paper proposes a novel method based on the dual-tree complex wavelet Transform (CWT). As CWT is effectively capable of representing edges in various directions, it is used to estimate the best intra angular mode in each coding block. The encoder then only analyzes a few modes around the best estimated mode. The experimental results show that searching only six modes around the estimated mode (estimated mode?±?3) plus DC and planar can predict the best mode in almost 80 % of the times. In fact the remaining unpredicted blocks have such unspecified behavior that even their random selections do not have any side effects. With all-intra configuration, searching with this method among the estimated mode?±?3 and DC and planar can improve the intra frame coding time by up to 4---10 times faster and can save the total encoding time over 32 %, while causing less than 0.4 % increase in bitrate and preserving the quality.

Analysis and Evaluation of HEVC

HEVC video coding standard is the next generation of H.264 coding standard, which is able to achieve twice coding efficiency compared with H.264/AVC High Profile. This paper analyzes the key features of HEVC standard based on HEVC reference software HM12.0. Experiments cover different resolutions ranging from 240p to 1600p. The most complex parts of HEVC decoder are intra prediction, loop filter and inverse transform when video sequences are coded with all-intra configuration. When video sequences are coded in random-access configuration, HEVC decoding process is occupied by motion compensation, which takes nearly half of the decoding time. Experiment results indicate that HEVC standard can achieve 50% average bit-rate saving compared with H.264.

Unified depth intra coding for 3D video extension of HEVC

With the development of high-efficiency video coding (HEVC), the newest video coding standard, 3D video extension of HEVC (3D-HEVC) has been actively investigated. Since 3D-HEVC uses multi-view texture and depth data for input, various coding tools have been added to HEVC. In 3D-HEVC, on top of the existing 35 HEVC intra modes, eight additional modes exist, which are specifically for depth coding. In this paper, we propose a unified depth intra coding method that incorporates such depth intra modes into the regular intra mode set. In particular, the most rarely used HEVC intra modes are replaced by depth intra modes. As a result, binarization for depth intra modes is removed. Furthermore, the most probable mode selection procedure is modified to consider the elimination of several angular intra modes. The proposed method is implemented and tested on 3D video HEVC test model version 7.0. Simulation results report 2.2 % synthesis gain under all-intra configuration.