“…Additionally, Table II presents the results of two related works. The solution proposed in [5] obtained average BDBR increase of 0.19% and 16.5% of ETR. The work [6] attained average ETR increase of 35.4%, with a 1.06% increase in the BDBR.…”
Section: Resultsmentioning
confidence: 96%
“…Some works already have proposed solutions to reduce the intra coding time of depth maps. Fu et al [5] propose an algorithm that uses corner detection for skipping unnecessary DMM evaluations on blocks where the conventional HEVC intra prediction modes can perform a good prediction. Wang et al [6] propose an algorithm that uses the Roberts operator, which uses local differences to detect edges in the image for skipping DMMs evaluation.…”
This paper presents a fast intra mode decision for depth map coding on 3D-High Efficiency Video Coding (3D-HEVC) based on decision trees. The proposed solution uses data mining and machine learning to correlate the encoder context attributes and build a set of decision trees. Each decision tree defines if a depth map block must be or not be evaluated by the Depth Modeling Modes (DMMs), considering the encoding context. The decision trees were trained using data extracted from the 3D-HEVC Test Model (3D-HTM) under all-intra encoder configuration. The proposed solution was evaluated according to the Common Test Conditions (CTC), reducing 50.2% the execution time of the depth map coding, and impacting only 0.07% in the Bjontegaard Delta BitRate (BDBR) of the synthesized views.
“…Additionally, Table II presents the results of two related works. The solution proposed in [5] obtained average BDBR increase of 0.19% and 16.5% of ETR. The work [6] attained average ETR increase of 35.4%, with a 1.06% increase in the BDBR.…”
Section: Resultsmentioning
confidence: 96%
“…Some works already have proposed solutions to reduce the intra coding time of depth maps. Fu et al [5] propose an algorithm that uses corner detection for skipping unnecessary DMM evaluations on blocks where the conventional HEVC intra prediction modes can perform a good prediction. Wang et al [6] propose an algorithm that uses the Roberts operator, which uses local differences to detect edges in the image for skipping DMMs evaluation.…”
This paper presents a fast intra mode decision for depth map coding on 3D-High Efficiency Video Coding (3D-HEVC) based on decision trees. The proposed solution uses data mining and machine learning to correlate the encoder context attributes and build a set of decision trees. Each decision tree defines if a depth map block must be or not be evaluated by the Depth Modeling Modes (DMMs), considering the encoding context. The decision trees were trained using data extracted from the 3D-HEVC Test Model (3D-HTM) under all-intra encoder configuration. The proposed solution was evaluated according to the Common Test Conditions (CTC), reducing 50.2% the execution time of the depth map coding, and impacting only 0.07% in the Bjontegaard Delta BitRate (BDBR) of the synthesized views.
“…To accelerate the evaluation of DMM, several methods have been proposed. The evaluation results of intra conventional modes and golden ratio search is proposed for fast DMM decision [10]. A feature corner point is raised to evaluatethe orientation of edges for decision of DMM skip [11].…”
Section: Compensated Prediction (Dcp) and Inter View Motion Predictiomentioning
The 3D extension of High Efficiency Video Coding (3D-HEVC) introduce Depth Modeling Mode (DMM) and 35 conventional intra modes to enhance the quality of coding, while bringing unacceptable computational complexity during the process of rough mode decision (RMD) and most probable mode (MPM). In this paper, we proposed a fast mode decision algorithm for texture map and depth map coding based on gradient information. Firstly, analyzing the characteristics of predict units (PU) in different intra mode applications that obtain the lowest RD-cost,then extracting the gradient information of PU to classify the PU into three types of gradient blocks and selecting appropriate candidate modes for the PU,thereby avoiding search each mode in coding process and skip the process of RMD and MPM early. Experimental results show that the proposed algorithm can achieve average 30.6% time saving with negligible reduction of coding performance.
“…In order to reduce the complexity of depth intra coding, many fast approaches have appeared [1]- [13]. They could be roughly divided into two categories: fast mode decision methods [1]- [9] and fast coding block partition methods [10]- [13]. In [1], an edge-based DMM skipping strategy was proposed in Hadamard transform domain.…”
Section: Introductionmentioning
confidence: 99%
“…In [4], the distortion calculation in DMM search is simplified as the squared Euclidean distance of variances (SEDV), and a probability-based early decision (PBED) method was proposed by studying the correlation between rough mode cost and final mode decision. Other algorithms for block level fast intra mode decision could be found in [5]- [9], which mainly focus on the acceleration of complex modes such as DMM. On the other hand, a coding block quad-tree pruning algorithm was designed in [10] by considering the variance of blocks and the estimated distortion of single depth intra mode [15].…”
The coding units (CU) partitioning in the 3D extension of the high efficiency video coding standard (3D-HEVC) is recursively conducted on different block sizes from 64 × 64 to 8 × 8. Besides, the depth coding in 3D-HEVC introduces several new coding tools for each CU to improve the coding efficiency, however, with great computational complexity. It is noted that only a small number of the CUs in recursive partitioning are encoded into the final bitstream. Among these CUs, the CUs with Intra 2N × 2N or Intra N × N as optimal modes have a very small proportion. In this paper, we thus propose an early determination of depth intra coding, where the coding stage of Intra 2N × 2N, Intra N × N or CU Splitting for the CUs could be early skipped based on several Decision Trees. Simulation results show that, with restrict the results from the tree leaves by different Gini thresholds, the proposed algorithm could save 41.14%-71.63% of the depth coding time with only a slight increase in BDBR. INDEX TERMS 3D-HEVC, CU size decision, decision tree, Gini thresholds, depth intra coding, fast mode decision.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.