• Journal of Computing Science and Engineering
• /
• v.4 no.3
• /
• pp.207-224
• /
• 2010

This paper presents a compression scheme for mesh geometry, which is suitable for mobile graphics. The main focus is to enable real-time decoding of compressed vertex positions while providing reasonable compression ratios. Our scheme is based on local quantization of vertex positions with mesh partitioning. To prevent visual seams along the partitioning boundaries, we constrain the locally quantized cells of all mesh partitions to have the same size and aligned local axes. We propose a mesh partitioning algorithm to minimize the size of locally quantized cells, which relates to the distortion of a restored mesh. Vertex coordinates are stored in main memory and transmitted to graphics hardware for rendering in the quantized form, saving memory space and system bus bandwidth. Decoding operation is combined with model geometry transformation, and the only overhead to restore vertex positions is one matrix multiplication for each mesh partition. In our experiments, a 32-bit floating point vertex coordinate is quantized into an 8-bit integer, which is the smallest data size supported in a mobile graphics library. With this setting, the distortions of the restored meshes are comparable to 11-bit global quantization of vertex coordinates. We also apply the proposed approach to compression of vertex attributes, such as vertex normals and texture coordinates, and show that gains similar to vertex geometry can be obtained through local quantization with mesh partitioning.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.11C
• /
• pp.1066-1070
• /
• 2006

In this paper, we present a residual filter to improve performance of H.264 compressed video. In general, noisy video sequences captured by imaging system result in decline of coding efficiency and unpleasant coding artifacts due to higher frequency components. By incorporating local statistics and quantization parameter into filtering process, the spurious noise is significantly attenuated and coding efficiency is improved, leading to improvement of visual quality and to bit-rate saving for given quantization step size. In addition, in order to reduce the complexity of the residual filter, noise induced by analyzing H.264 transformation and quantization processes are introduced. The simulation results show the capability of the proposed algorithm.

• Journal of Korea Multimedia Society
• /
• v.9 no.3
• /
• pp.296-306
• /
• 2006

We proposed the algorithm for the quantization noise reduction based on variable filter adaptive to edge signal in MPEG postprocessing system. In our algorithm, edge map and local modulus maxima in the decoded images are obtained by using 2D Mallat wavelet tilter. And then, blocking artifacts in inter-block are reduced by Gaussian LPF that is variable to filtering region according to edge map. Ringing artifacts in intra-block are reduced by 2D SAF according to local modulus maxima. Experimental results show that the proposed algorithm was superior to the conventional algorithms as regards PSNR, which was improved by 0.04-0.20 dB, and the subjective image quality.

• Journal of information and communication convergence engineering
• /
• v.16 no.3
• /
• pp.142-147
• /
• 2018

We present a practical design algorithm for quantizers at nodes in distributed systems in which each local measurement is quantized without communication between nodes and transmitted to a fusion node that conducts estimation of the parameter of interest. The benefits of vector quantization (VQ) motivate us to incorporate the VQ strategy into our design and we propose a low-complexity design technique that seeks to assign vector codewords into sets such that each codeword in the sets should be closest to its associated local codeword. In doing so, we introduce new distance metrics to measure the distance between vector codewords and local ones and construct the sets of vector codewords at each node to minimize the average distance, resulting in an efficient and independent encoding of the vector codewords. Through extensive experiments, we show that the proposed algorithm can maintain comparable performance with a substantially reduced design complexity.

• Journal of Broadcast Engineering
• /
• v.15 no.5
• /
• pp.672-680
• /
• 2010

This paper presents an adaptive quantization matrix selection scheme for H.264 video encoding. Conventional H.264 coding standard applies the same quantization matrix to the entire video sequence without considering local characteristics in each frame. In this paper, we propose block adaptive selection of quantization matrix according to edge directivity of each block. Firstly, edge directivity of each block is determined using intra prediction modes of its spatially adjacent blocks. If the block is decided as a directional block, new weighted quantization matrix is applied to the block. Otherwise, conventional quantization matrix is used for quantization of the non-directional block. Since the proposed weighted quantization is designed based on statistical distribution of transform coefficients in accordance with intra prediction modes, we can achieve high coding efficiency. Experimental results show that the proposed scheme can improve coding efficiency by about 2% in terms of BD bit-rate.

• Proceedings of the IEEK Conference
• /
• 2000.11d
• /
• pp.155-158
• /
• 2000

In this paper, the quantization noise in block-based video coding is analyzed, and a post-processing method based on the analysis is presented for reducing the quantization noise by using a wavelet transform(WT). In the proposed method, the quantization noise is considered as the sum of a blocking noise expressed as a deterministic profile and the random remainder noise. Each noise is removed in a viewpoint of image restoration using a 1-D WT, which yields a regularized differentiation. The blocking noise first is reduced by weakening the strength of each blocking noise component that appears as an impulse in the first scale wavelet domain. The impulse strength estimation is performed using median filter, quantization parameter(QP), and local activity. The remainder noise, which is considered as a white noise at non-edge pixels, then is reduced by soft-thresholding. The experimental results show that the proposed method yields better performance in terms if subjective quality as well as PSNR performance over VM post-filter in MPEG-4 for all test sequences of various compression ratios. We also present a fast post-processing in spatial domain equivalent to that in wavelet domain for real-time application.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.2C
• /
• pp.41-48
• /
• 2005

In this paper, we present a low complexity modified Gaussian model based pre-processing filter to improve the performance of H.264 compressed video. Video sequence captured by general imaging system represents the degraded version due to the additive noise which decreases coding efficiency and results in unpleasant coding artifacts due to higher frequency components. By incorporating local statistics and quantization parameter into filtering process, the spurious noise is significantly attenuated and coding efficiency is improved for given quantization step size. In addition, in order to reduce the complexity of the pre-processing filter, the simplified local statistics and quantization parameter are introduced. The simulation results show the capability of the proposed algorithm.

• KIEE International Transaction on Systems and Control
• /
• v.12D no.1
• /
• pp.7-11
• /
• 2002

In this study, fur the shape-based image retrieval, a method using local differential invariants is proposed. This method calculates the differential invariant feature vector at every feature point extracted by Harris comer point detector. Then through vector quantization using LBG algorithm, all feature vectors are represented by a codebook index. All images are indexed by the histogram of codebook index, and by comparing the histograms the similarity between images is obtained. The proposed method is compared with the existing method by performing experiments for image database including various 1100 trademarks.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.4
• /
• pp.848-854
• /
• 1996

Many image display devices allow only a limited number of colors to be simultaneously displayed. in disphaying of natural color image using color palette, it is necessary to construct an optimal color palette and the optimal mapping of each pixed of the original image to a color from the palette. In this paper, we proposed the clustering algorithm using local region block centered one color cluster in the prequantized 3-D histogram. Cluster pairs which have the least distortion error are merged by considering distortion measure. The clustering process is continued until to obtain the desired number of colors. The same as the clustering process, original color value. The proposed algorithm incroporated with a spatial activity weighting value which is reflected sensitivity of HVS quantization errors in smoothing region. This method produces high quality display images and considerably reduces computation time.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.662-663
• /
• 2020

최근 MPEG 에서는 인공신경망 모델을 다양한 딥러닝 프레임워크에서 상호운용 가능한 포맷으로 압축 표현할 수 있는 NNR(Compression of Neural Network for Multimedia Content Description and Analysis) 표준화를 진행하고 있다. 본 논문에서는 MPEG-NNR 에서 CNN 모델을 압축하기 위한 지역 비선형 양자화(Local Non-linear Quantization: LNQ) 기법을 제시한다. 제안하는 LNQ 는 균일 양자화된 CNN 모델의 각 계층의 가중치 행렬 블록 단위로 추가적인 비선형 양자화를 적용한다. 또한, 제안된 LNQ 는 가지치기(pruning)된 모델의 경우 블록내의 영(zero) 값의 가중치들은 그대로 전송하고 영이 아닌 가중치만을 이진 군집화를 적용한다. 제안 기법은 음성 분류를 위한 CNN 모델(DCASE Task)의 압축 실험에서 기존 균일 양자화를 대비 동일한 분류 성능에서 약 1.78 배 압축 성능 향상이 있음을 확인하였다.