Evaluation of coding efficiency is traditionally modeled as a continuous rate-distortion (R-D) function, where the peak signal-to-noise ratio (PSNR) adopted quality measure. Although PSNR-versus-bitrate curve offers some useful tradeoff information between video and bit-rates, it does not take human perceptual experience into account. In this work, by following recent image/video assessment framework based on just-noticeable-difference (JND) notion, we conduct subjective test for HEVC (High...

While adaptive splitting the quadtree-structured coding units according to video content is utmost important in achieving best encoding performance of High Efficiency Video Coding (MPEG-H HEVC/H.265), newest standard, its optimal decision requires much computation. To resolve this complexity problem, we extend keypoint-based CU depth (KCD) method by further providing it with capability reduce more. Experimental results show that can save time 11.6% sacrificing BD-rate only 0.68% compared...

A deblocking filtering method that reduces the blocking artifacts result from use of illumination change-adaptive motion compensation method, which was already adopted in current multiview video coding (MVC) standard, is introduced. When macroblock (MB)-based used to compensate local changes sequences, neighboring MBs could have different values represent average luminance MBs, horizontally and vertically. This phenomenon causes horizontal/vertical on MB boundaries due difference between...

We present a novel video codec for supporting entertainment-quality video. It has new coding tools such as an intra prediction with offset, integer sine transform, and enhanced block-based adaptive loop filter. These are used adaptively in the processing of prediction, filtering. In our experiments, proposed achieved average reduction 13.35% BD-rate relative to H.264/AVC 720p sequences.

An alternative transform based on the correlation of residual signal is proposed for improvement H.264/AVC coding efficiency. To make use this in video coding, discrete sine and integer are used alternately with cosine order to greatly compact energy when coefficients relatively low. Therefore, suggested be conjunction H.264/AVC. The that selects optimal between two transforms by using rate-distortion optimization shows a gain compared Based experimental results, method achieves BD-PSNR up...

In this letter, we present an enhanced block-based adaptive loop filter (E-BALF) with multiple symmetric structures. The E-BALF adapts various structures in a rate-distortion optimization sense, reflecting the statistical properties of each image video sequence. Experimental results show that proposed method achieves reduction Bjøntegaard delta (BD)-bitrate by average 9.60% compared Joint Model 11.0 H.264/AVC. Compared to state-of-the-art BALF, up 1.13% BD-bitrate is achieved.

We propose a rate-distortion optimized transform coding method that adaptively employs either integer cosine is an integer-approximated version of discrete (DCT) or sine (IST) in sense. The DCT has been adopted most video-coding standards known as suboptimal substitute for the Karhunen-Loève transform. However, according to correlation signal, alternative can achieve higher efficiency. introduce (DST) achieves high-energy compactness coefficient range −0.5 0.5 and applied current design...

Self-similarity based SR method on the LF-HF domain basically relies assumption that LF patch is highly correlated with HF patch. However, this content-specific significantly, and we can occasionally observe little correlation between especially for texture region of natural images. In paper propose a new self-similarity to reflect observation. There are two differences proposed existing methods. First, details target HR image recovered by finding similar patches MF in case region. Second,...

This paper proposes a Subsampled Sum-Modified-Laplacian (SSML) operator for the block classification of Adaptive Loop Filter (ALF) in Versatile Video Coding (VVC). The VVC Test Model (VTM)-2.0 includes Geometry transformation-based ALF (GALF) with 4 × classification, single 7 Luma diamond-shaped filter, and spatial adaptation at Tree Block (CTB) level to improve coding efficiency VVC. However, 1-D (1-Dimensional) ModifiedLaplacian (ML) values various directions are calculated all sample...

Video compression exploits statistical, spatial, and temporal redundancy, as well transform quantization. In particular, the in a frequency domain plays major role energy compaction of spatial data into data. The high efficient video coding standard uses type-II discrete cosine (DCT-II) type-VII sine (DST-VII) to improve efficiency residual However, DST-VII is applied only Intra 4 × block because it yields relatively small gains larger than block. this study, after rearranging block, we...

To improve coding efficiency by exploiting the local inter-component redundancy between luma and chroma components, cross-component linear model (CCLM) is included in versatile video (VVC) standard. In CCLM mode, parameters are derived from neighboring samples of current block. Furthermore, predicted reconstructed collocated block with parameters. However, as design VVC test (VTM)-6.0 has many conditional branches its processes to use only available samples, implementation parallel...

A sophisticated video surveillance system involves the problem of limited storage to record data for long time. Video compression technology is an effective solution address this problem. Inspired by success neural network-based approaches in computer vision, research on coding has emerged. With aim achieving improved efficiency, investigation inter prediction plays a crucial role coding. In paper, we propose convolutional network (CNN)-based generation and enhancement method (GEIP)...

본 논문에서는 참조 프레임 혹은 시간적으로 이전에 부호화한 프레임을 통해 현재 프레임의 LCU 분할구조를 예측하여 부호화하는 방법을 제안한다. HEVC에서는 CU로 부호화 및 복호화를 수행하는데, CU의 기본이 되는 단위로 영상의 특성에 따라 결정하여 영상을 적응적으로 부호화한다. 이 때, 부호화하려는 LCU의 분할구조와 내의 동일한 위치에 대응되는 LCU(Co-located LCU)의 분할구조는 매우 유사한 특성이 있다. 따라서 인코더의 복잡도를 낮추기 위하여 결정할 Co-located 복잡성을 분할구조 정보를 예측하고 분할구조에 포함될 확률이 높은 CU만 제안 방법의 시뮬레이션 결과로서, 인코더만을 변경하여 인코더 낮추는 방법이 기존 대비 복잡도가 평균 21.3% 감소하였고, 디코더 복잡도는 거의 비슷했으며, BD-Bitrate는 최대 0.6% 증가하였다. 또한 인코더에서 때 분할 부호화하고, CU 복호화하는 과정을 변경하는 BD-Bitrate를 감소시키는 제안하였다....

본 논문은 H.264/AVC 화면내 부호화에서 부호화 효율 향상을 위해 율-왜곡 최적화를 이용한 예측 오프셋 보상 방법을 제안한다. H.264/AVC의 부호화는 주변 블록의 복원 화소들을 현재 부호화하려는 블록으로 활용함으로써 공간적 상관성을 제거하고 효율을 향상시킨다. 제안 방법은 부호화의 오차를 감소시키기 위해, 비용 관점에서의 최적 값을 갖는 오프셋을 기존 블록에 더하여 정확도를 높인다. 따라서 오차 신호의 양자화된 변환 계수를 감소시키며 원본 블록과 복원된 블록 간의 왜곡을 감소시켜 성능을 향상시킬 수 있다. 추가적으로, 휘도 성분의 성능 보상이 사용되는 색차 성분에 JM 11.0에서 화면간 계수 임계 처리 적용한다. 논문의 실험에서는 제안하는 방법이 11.0과 비교 실험했을 때 High Profile 환경에서 평균 2.45%의 비트율 감소와 최대 4.41%의 비트율을 감소시킬 있음을 보인다. In this paper, an intra prediction offset...

In video compression, transformation plays a significant role in the energy compaction of spatial domain data into frequency data. HEVC intra prediction, Discrete Cosine Transform (DCT) and Sine (DST) are used to concentrate residual signals low-frequency components. DCT has shown good compression performance both inter coding, but when not uniformly distributed, its coding efficiency decreases. This paper proposes transform method that applies or residual-rearranged DST improve coding. The...

The High Efficiency Video Coding (MPEG-H HEVC/ITU-T H.265) is the newest video coding standard which has quadtree-structured unit (CU). quadtree-structure splits a CU adaptively, and its optimum depth can be determined by rate-distortion optimization. Such HEVC encoding requires very high computational complexity for decision. Motivated that blob detection, well-known algorithm in computer vision, detects keypoints pictures decision of needs to consider frequency energy distribution, this...

본 논문에서는 현재 표준화가 진행중인 HEVC(High Efficiency Video Coding)의 고정 비트율(CBR: Constant Bitrate) 부호화를 위한 비트율 제어(rate control) 기법을 다룬다. 기존의 H.264/AVC의 제어 기법에서는 계층적 부호화 구조의 특성을 고려하지 못한 사용하여 정확한 제어에 어려움이 있다. HEVC의 참조 소프트웨어인 HM(HEVC test Model)에는 CBR 알고리즘이 제공되지 않은 상태이며, 추후 모듈이 추가될 것으로 예상된다. 이때 HM의 공통실험조건에 정의된 임의접근(Random Access: RA) 모드의 구조에서의 보다 제어가 요구된다. 따라서, 효율적인 제어를 위해 2차 비트율-왜곡(Rate-Distortion: R-D) 모델 기반의 시간계층 및 프레임 타입에 따른 반영한 제시한다. 또한, 단위에 비트 변동율과 화질 손실 간에 발생하는 상충관계를 고려하여 단위를 단위와 기본 단위(Coding Unit:...

We present a novel video codec for supporting entertainment-quality. It has new coding tools that are intra prediction with offset, integer sine transform adaptive quantization matrix selection, uni-directional prediction, and enhanced block-based loop filter. These used adaptively in processing of transform, filtering. In experiments, the proposed achieved high rate reduction by average 14.77% BD-rate relative to H.264/AVC 720p sequences. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...

In this paper, an alternative transform based on the correlation of residual block is proposed for improvement H.264/AVC coding efficiency. A discrete sine used alternately with a cosine in order to greatly compact energy signal when coefficients are range -0.5 0.5. Therefore, suggested be conjunction H.264/AVC. The selecting optimal between two transforms by using rate-distortion optimization shows gain compared method achieves PSNR up 1.0dB JM 10.2 at relatively high bitrates.

This paper presents an efficient way of 2D reprojection method 3D voxel data which is generated from multiple image sources. In order to minimize the quality degradation by postprocessing such as hole filling, A using gaussian aggregation attribute intensity proposed. Proposed efficiently removes generation holes while maintaining graceful projected planes, can finally reconstruct original source images with less degradation.