A simple image coder based on triangle meshes is proposed, and several mesh-generation methods are presented, some previously proposed and some new. Then, our coder is used to evaluate the performance of the various m...
详细信息
A simple image coder based on triangle meshes is proposed, and several mesh-generation methods are presented, some previously proposed and some new. Then, our coder is used to evaluate the performance of the various mesh-generation methods. One of our proposed mesh-generation schemes, inspired by the work of Garland and Heckbert, is shown to perform best. The performance of linear and Clough-Tocher interpolation is also compared. Linear interpolation is found to perform better than Clough-Tocher interpolation, with the latter performing more poorly largely due to overshoot/undershoot near image edges.
The context-based adaptive variable length coding (CAVLC) is a new and efficient entropy coding tool for the H.264/AVC. Although the CAVLC provides the excellent compression ratio, the computational complexity of the ...
详细信息
The context-based adaptive variable length coding (CAVLC) is a new and efficient entropy coding tool for the H.264/AVC. Although the CAVLC provides the excellent compression ratio, the computational complexity of the CAVLC decoder (CAVLD) is higher than that of the traditional variable length decoder. In this paper, we propose a low-power and low-cost architecture of the CAVLC decoder for the H.264/AVC baseline profile. The research derives the optimum power model for the variable length look-up table (LUT) of the CAVLC decoder, and then we divide the decoding phase of the LUT into two decoding layers. We also merge the common code words to reduce the hardware cost among the different LUTs in the second decoding layer. Moreover, the design is based on the 0.18-mum TSMC CMOS technology. The experimental results show that the proposed decoder operates at the 125 MHz clock frequency with the hardware cost of 4412 gates. Furthermore, the proposed design can reduce the power consumption about 44% to 48% more than the previous low-power CAVLD schemes do.
JPEG2000 will be the international standard to continuous-tone still image coding in next generation. JPEG2000 algorithm is different from JPEG algorithm in filtering and entropy coding. In this thesis, we propose DWT...
详细信息
ISBN:
(纸本)9781424425983
JPEG2000 will be the international standard to continuous-tone still image coding in next generation. JPEG2000 algorithm is different from JPEG algorithm in filtering and entropy coding. In this thesis, we propose DWT filter hardware design to improve performance. Hardware module which performs the DWT operation is designed by Verilog-HDL and synthesised by ISE 6.3i. As a result of implementing DWT hardware, a speed gain of about 35 times is achieved when a process did DWT by software.
In this paper, we propose the efficient NAL (Network Abstraction Layer) decoder for H.264/AVC baseline profile which is specified for mobile telecommunication. Therefore, we can simplify NAL decoder for baseline profi...
详细信息
ISBN:
(纸本)9781424425983
In this paper, we propose the efficient NAL (Network Abstraction Layer) decoder for H.264/AVC baseline profile which is specified for mobile telecommunication. Therefore, we can simplify NAL decoder for baseline profile. Many parameters and complex algorithms are fixed or removed in baseline profile. This optimization approach can be defined by target based NAL decoder implementation. We discuss about this target based NAL decoder and implementation especially in H.264/AVC baseline profile.
This paper proposes an improved lossless compression algorithm based on the spatial-spectral hybrid *** choose the prediction modes between the spatial and the spectral domains by computing the local correlation *** s...
详细信息
This paper proposes an improved lossless compression algorithm based on the spatial-spectral hybrid *** choose the prediction modes between the spatial and the spectral domains by computing the local correlation *** such coefficient is larger than the pre-designed threshold,the spectral linear predictor is adopted,which is able to capture more spectral correlation by re-estimating the ***,MED predictor is ***,prediction error images are coded by RICE *** are carried out on AVIRIS *** results show that the proposed method outperforms 3D-CALIC algorithm,MED and GAP spatial lossless prediction algorithms.
In traditional modes of lossy compression, attaining low distortion letter-by-letter on a vector of source letters X 1 N =(X 1 , X 2 ,..., X N )isinRopf N is the implicit aim. We consider here instead the goal of est...
详细信息
In traditional modes of lossy compression, attaining low distortion letter-by-letter on a vector of source letters X 1 N =(X 1 , X 2 ,..., X N )isinRopf N is the implicit aim. We consider here instead the goal of estimating at the destination a function G(X 1 N ) of the source data under the constraint that each X i must be separately scalar quantized. The design of optimal fixed- and variable-rate scalar quantizers is considered under the assumptions of high-resolution quantization theory, yielding optimal point densities for regular quantizers. Additionally, we consider how performance scales with N for certain classes of functions. This demonstrates potentially large improvement from consideration of G in the quantizer design.
It was recently shown that the symmetric multiple-description (MD) quadratic rate-distortion function for memoryless Gaussian sources and two descriptions can be achieved by dithered delta-sigma quantization combined ...
详细信息
It was recently shown that the symmetric multiple-description (MD) quadratic rate-distortion function for memoryless Gaussian sources and two descriptions can be achieved by dithered delta-sigma quantization combined with memoryless entropy coding. In this paper, we generalize this result to stationary (colored) Gaussian sources by combining noise shaping and source prediction. We first propose a new representation for the test channel that realizes the MD rate-distortion function of a Gaussian source, both in the white and in the colored source case. We then show that this test channel can be materialized by embedding two source prediction loops, one for each description, within a common noise shaping loop. While the noise shaping loop controls the tradeoff between the side and the central distortions, the role of prediction (like in differential pulse code modulation) is to extract the source innovations from the reconstruction at each of the side decoders, and thus reduce the coding rate. Finally, we show that this scheme achieves the MD rate-distortion function at all resolutions and all side-to-central distortion ratios, in the limit of high dimensional quantization.
In this paper, a novel intra coding scheme is proposed. The proposed scheme improves H.264 intra coding from three aspects: 1) H.264 intra prediction is enhanced with additional bi-directional intra prediction modes; ...
详细信息
In this paper, a novel intra coding scheme is proposed. The proposed scheme improves H.264 intra coding from three aspects: 1) H.264 intra prediction is enhanced with additional bi-directional intra prediction modes; 2) H.264 integer transform is supplemented with directional transforms for some prediction modes; and 3) residual coefficient coding in CAVLC is improved. Compared to H.264, together the improvements can bring on average 7% and 10% coding gain for CABAC and for CAVLC, respectively, with average coding gain of 12% for HD sequences.
Microsoft's recently proposed new image compression codec HDPhoto is currently undergoing ISO standardization as JPEG-XR. Even though performance measurements carried out by the JPEG committee indicated that the P...
详细信息
Microsoft's recently proposed new image compression codec HDPhoto is currently undergoing ISO standardization as JPEG-XR. Even though performance measurements carried out by the JPEG committee indicated that the PSNR performance of HDPhoto is competitive, the visual performance of HDPhoto showed notable deficits, both in subjective and objective tests. This paper introduces various techniques that improve the visual performance of HDPhoto without leaving the current codestream definition. Objective measurements performed by the author indicate that the modified encoder, while staying backwards compatible to the current standard proposition, improves visual performance significantly, and the performance of the modified encoder is similar to JPEG.
H.264 is the latest international video coding standard. It is a joint work by the ITUT Video coding Experts Group and the ISO/IEC Moving Picture Experts Group. H.264 contains a number of new features which make it co...
详细信息
H.264 is the latest international video coding standard. It is a joint work by the ITUT Video coding Experts Group and the ISO/IEC Moving Picture Experts Group. H.264 contains a number of new features which make it compress video sequences much more effectively than previous standards and can provide more flexibility for application to a wide variety of network environments. But the H.264 standard also brings much more computations, especially when the rate-distortion optimization function is on. In order to reduce the complexity of rate-distortion cost of the rate part, we propose al rate estimation methods to avoid doing the real entropy coding method. The estimation method proposed is based on the notification of CAVLC tables, and is changed adaptively according to total_coeff. The final result shows that the proposed estimation method can reduce the encoding computation by 50% through the intra coding with ignorable degradation of coding performance.
暂无评论