We discuss the applications of algebraic statistics to fractional factorial design with special emphasis on the choice of level coding. In particular, we deal with the theory of Bayley's level codings in that fram...
详细信息
We discuss the applications of algebraic statistics to fractional factorial design with special emphasis on the choice of level coding. In particular, we deal with the theory of Bayley's level codings in that framework. (c) 2007 Elsevier B.V. All rights reserved.
After the development of the next generation video coding standard, referred to as high efficiency video coding (HEVC), the joint collaborative team of the ITU-T video coding experts group and the ISO/IEC moving pictu...
详细信息
After the development of the next generation video coding standard, referred to as high efficiency video coding (HEVC), the joint collaborative team of the ITU-T video coding experts group and the ISO/IEC moving picture experts group has now also standardized a lossless extension of such a standard. HEVC was originally designed for lossy video compression, thus, not ideal for lossless video compression. In this paper, we propose an efficient residual data coding method for HEVC lossless video compression. Based on the fact that there are statistical differences of residual data between lossy and lossless coding, we improved the HEVC lossless coding using sample-based angular prediction (SAP), modified level binarization, and binarization table selection with the weighted sum of previously encoded level values. Experimental results show that the proposed method provides high compression ratio up to 11.32 and reduces decoding complexity.
This paper presents a high performance design for Context-Based Adaptive Variable Length coding (CAVLC) used in the H.264/AVC standard. A two-stage encoder is proposed to make the scan and encode stage work simultaneo...
详细信息
This paper presents a high performance design for Context-Based Adaptive Variable Length coding (CAVLC) used in the H.264/AVC standard. A two-stage encoder is proposed to make the scan and encode stage work simultaneously. The scan engine scans four coefficients at each cycle. Parallel encoder for four "levels" and parallel encoder for four "Run-before" are adopted to accelerate the encode engine. Only 120 cycles at most are needed to process one MB. The proposed CAVLC encoder can support 4 Kx2 K@60 fps (frame per second) real-time encoding at 250MHz and the gate count is about 32 k.
In the emerging high efficiency video coding (HEVC) standard, a Golomb-Rice code is adopted to binarize level information. The Golomb-Rice code is optimal for encoding symbols with the exponential probability distribu...
详细信息
In the emerging high efficiency video coding (HEVC) standard, a Golomb-Rice code is adopted to binarize level information. The Golomb-Rice code is optimal for encoding symbols with the exponential probability distribution. In general, quantized transform coefficients of natural images show the exponential probability distribution. However, screen contents usually have a totally different probability distribution, and the Golomb-Rice code is not appropriate for screen content coding. In this paper, we focus on a new entropy coding scheme for screen content coding. In the proposed scheme, we clip inefficient high magnitude coefficient levels that do not fit the exponential probability distribution. Then, we apply a limited length Golomb-Rice code to binarize clipped levels. From experimental results, it is verified that the proposed method achieves on average 0.60 % BD-rate saving and up to 1.13 % BD-rate saving, compared to HEVC screen content coding. When the proposed method is combined with a well-known screen content coding technique, transform skipping, it shows up to 24.02 % BD-rate saving.
In H.264/AVC and the variants, the coding of context-based adaptive variable length codes (CAVLC) requires demanding operations, particularly at high bitrates such as 100 Mbps. This letter presents two approaches to a...
详细信息
In H.264/AVC and the variants, the coding of context-based adaptive variable length codes (CAVLC) requires demanding operations, particularly at high bitrates such as 100 Mbps. This letter presents two approaches to accelerate the coding operation substantially. Firstly, in the architectural aspect, we propose component-level parallelism and pipeline techniques capable of processing high-bitrate video data in a macroblock(MB)-level pipelined codec architecture. The second approach focuses on a specific part of the coding process, i.e., the residual block coding, in which the coefficient levels are coded without using look-up tables so we minimize the pertaining logic depth in the critical path, and we achieve higher operating clock frequencies. Additionally, two coefficient levels are processed in parallel by exploiting a look-ahead technique. The resulting architecture, merged in the NIB-level pipelined codec system, is capable of coding up to 100 Mbps bitstreams in real-time, thus accommodating the real-time encoding of 1080p@60 Hz video.
Sound level processing is a fundamental function of the auditory system. To determine how the cortex represents sound level, it is important to quantify how changes in level alter the spatiotemporal structure of corti...
详细信息
Sound level processing is a fundamental function of the auditory system. To determine how the cortex represents sound level, it is important to quantify how changes in level alter the spatiotemporal structure of cortical ensemble activity. This is particularly true for echolocating bats that have control over, and often rapidly adjust, call level to actively change echo level. To understand how cortical activity may change with sound level, here we mapped response rate and latency changes with sound level in the auditory cortex of the pallid bat. The pallid bat uses a 60-30 kHz downward frequency modulated (FM) sweep for echolocation. Neurons tuned to frequencies between 30 and 70 kHz in the auditory cortex are selective for the properties of FM sweeps used in echolocation forming the FM sweep selective region (FMSR). The FMSR is strongly selective for sound level between 30 and 50 dB SPL. Here we mapped the topography of level selectivity in the FMSR using downward FM sweeps and show that neurons with more monotonic rate level functions are located in caudomedial regions of the FMSR overlapping with high frequency (50-60 kHz) neurons. Non-monotonic neurons dominate the FMSR, and are distributed across the entire region, but there is no evidence for amplitopy. We also examined how first spike latency of FMSR neurons change with sound level. The majority of FMSR neurons exhibit paradoxical latency shift wherein the latency increases with sound level. Moreover, neurons with paradoxical latency shifts are more strongly level selective and are tuned to lower sound level than neurons in which latencies decrease with level. These data indicate a clustered arrangement of neurons according to monotonicity, with no strong evidence for finer scale topography, in the FMSR. The latency analysis suggests mechanisms for strong level selectivity that is based on relative timing of excitatory and inhibitory inputs. Taken together, these data suggest how the spatiotemporal spread of
The dependence of fMRI activation on sound level was examined throughout the auditory pathway of normal human listeners using continuous broadband noise, a stimulus widely used in neuroscientific investigations of aud...
详细信息
The dependence of fMRI activation on sound level was examined throughout the auditory pathway of normal human listeners using continuous broadband noise, a stimulus widely used in neuroscientific investigations of auditory processing, but largely neglected in neuro-imaging. Several specialized techniques were combined here for the first time to enhance detection of brainstem activation, mitigate scanner noise, and recover temporal resolution lost by the mitigation technique. The main finding was increased activation with increasing level in cochlear nucleus, superior olive, inferior colliculus, medial geniculate body and auditory cortical areas. We suggest that these increases reflect monotonically increasing activity in a preponderance of individual auditory neurons responsive to broadband noise. While the time-course of activation changed with level, the change was subtle and only significant in a part of the cortex. To our knowledge, these are the first fMRI data showing the effects of sound level in subcortical centers or for a non-tonal, non-speech stimulus at any stage of the pathway. The present results add to the body of parametric data in normal human listeners and are fundamental to the design of any fMRI experiment employing continuous noise. (c) 2006 Elsevier B.V. All rights reserved.
This paper presents a high performance VLSI design of Context-Based Adaptive Variable Length-coding (CAVLC) for 4096x2160p@60fps H.264/AVC encoder. A parallel architecture is proposed to make the scan and encode stage...
详细信息
ISBN:
(纸本)9781467302197
This paper presents a high performance VLSI design of Context-Based Adaptive Variable Length-coding (CAVLC) for 4096x2160p@60fps H.264/AVC encoder. A parallel architecture is proposed to make the scan and encode stage work simultaneously. Four coefficients are scanned in parallel, and four levels and Run_before are coded in parallel. From experimental results, only 120 cycles at most are needed to process one macroblock (MB), which reduce more than 50% cycles compare to state-of-the-art designs. The hardware implementation results show that the proposed design achieves real-time encoding at 250 MHz and the hardware cost is about 32k gates.
暂无评论