2014 19th International Conference on Digital Signal Processing最新文献_第7页

A low complexity and high performance interpolation filter for MPEG IVC 一个低复杂度和高性能的mpegivc插值滤波器

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900817

Hao Lv, Ronggang Wang, Yangang Cai, Huizhu Jia, Xiaodong Xie, Wen Gao

{"title":"A low complexity and high performance interpolation filter for MPEG IVC","authors":"Hao Lv, Ronggang Wang, Yangang Cai, Huizhu Jia, Xiaodong Xie, Wen Gao","doi":"10.1109/ICDSP.2014.6900817","DOIUrl":"https://doi.org/10.1109/ICDSP.2014.6900817","url":null,"abstract":"Fractional-pel motion compensation is widely adopted in the modern video coding standards such as H.264/AVC, AVS, and HEVC etc. The interpolation filter is a critical factor that influences the coding efficiency. In this paper, a generation algorithm of interpolation filter coefficients is utilized. Based on the coefficients generation algorithm, three different tap filters, namely 6 tap, 8 tap, 10tap, are tested. A combination of 6 tap and 8 tap interpolation filters is proposed and proved to be optimal scheme considering both performance and computational complexity. And it is beneficial to make software optimization more effective, especially when SIMD-like (Single Instruction Multiple Data) operation is used. Experiments show that the average BD-rate gains on luma Y, chroma U and V are 8.01%, 5.08% and 4.98% for CS1 (Constraint set 1), and 9.21%, 7.53% and 7.63% for CS2 (Constraint set 2) in MPEG IVC reference software ITM5.0, compared with the traditional IVC interpolation method. The coding efficiency gains are significant for some video sequences and can reach up to 28.7%. With the merits of high performance and low complexity, our proposed method is formally adopted by MPEG IVC.","PeriodicalId":301856,"journal":{"name":"2014 19th International Conference on Digital Signal Processing","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114642953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Green noise video halftoning 绿噪视频半调

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900807

Yik-Hing Fung, Y. Chan

引用次数: 2

Novel CAVLC design for secondary SP-frame 新颖的二次sp车架CAVLC设计

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900833

Hongbin Zhang, C. Fu, W. Su, Yui-Lam Chan

引用次数: 1

Multi-resolution PR NMDFBs for programmable variable bandwidth filter in wideband digital transceivers 用于宽带数字收发器可编程可变带宽滤波器的多分辨率PR nmdfb

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900809

F. Harris, Behrokh Farzad, E. Venosa, Xiaofei Chen

{"title":"Multi-resolution PR NMDFBs for programmable variable bandwidth filter in wideband digital transceivers","authors":"F. Harris, Behrokh Farzad, E. Venosa, Xiaofei Chen","doi":"10.1109/ICDSP.2014.6900809","DOIUrl":"https://doi.org/10.1109/ICDSP.2014.6900809","url":null,"abstract":"This paper describes a novel application of perfect reconstruction (PR) non-maximally decimated filter banks (NMDFBs). PR non-maximally decimated analysis/synthesis chains are used for performing resampling, channelization and filtering in very wideband software defined radios. Digital wideband filtering is currently limited by the hardware clock speed which constrains the maximum signal bandwidth that can be processed digitally. Polyphase channelizers decrease the sample rate of the input signals and parallelize the wideband filtering process at cost of a single filter. In this paper we show how to assemble multiple-tier PR analysis/synthesis channelizer chains for implementing, in a very efficient way, variable bandwidth digital filters and multi-resolution systems operating on very wideband signals. When an internal tier analysis channelizer is applied to a single output channel of a first external tier analysis channelizer it decomposes this channel in multiple narrower subchannels and, by applying appropriate complex processing elements to those sub-channels we can modify the designed filter bandwidth as desired.","PeriodicalId":301856,"journal":{"name":"2014 19th International Conference on Digital Signal Processing","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123458030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 7

Corticomuscular coherence analysis on the static and dynamic tasks of hand movement 手部运动静态和动态任务的皮质肌肉一致性分析

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900757

Anshuang Fu, Rui Xu, Feng He, Hongzhi Qi, Lixin Zhang, Dong Ming, Yanru Bai, Zhiguo Zhang

{"title":"Corticomuscular coherence analysis on the static and dynamic tasks of hand movement","authors":"Anshuang Fu, Rui Xu, Feng He, Hongzhi Qi, Lixin Zhang, Dong Ming, Yanru Bai, Zhiguo Zhang","doi":"10.1109/ICDSP.2014.6900757","DOIUrl":"https://doi.org/10.1109/ICDSP.2014.6900757","url":null,"abstract":"The synchronization between cortical motor and muscular activity can be revealed by corticomuscular coherence (CMC). This paper designed two neuromuscular activity paradigms of hand movement, i.e. static gripping task and dynamic finger moving task. The electroencephalography (EEG) from C3 and C4 channels and the surface electromyography (sEMG) from the flexor digitorum superficialis were collected simultaneously from 4 male and 4 female right-handed healthy young subjects. For the static griping task, CMCs during low-level forces under 4%, 8%, and 16% MVC (Maximal Voluntary Contraction) were investigated by using magnitude squared coherence calculated from EEGs and sEMGs. For the dynamic finger moving task, the time-frequency domain analysis was used to process dynamic data of temporary action in a period of 2 seconds and get the latency of the maximum CMC. The results of this study indicated that the force increasing within the low-level range in static task is associated with the enhanced CMC. The maximum amplitude of CMC occurred about 0.3-0.5s after the onset of hand movement. Subjects showed significant CMC performance both in static and dynamic task of hand movement.","PeriodicalId":301856,"journal":{"name":"2014 19th International Conference on Digital Signal Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128803725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 5

Signal classification based on block-sparse tensor representation 基于块稀疏张量表示的信号分类

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900687

S. Zubair, Wenwu Wang

引用次数: 2

Parallelized feature extraction and acoustic model training 并行特征提取和声学模型训练

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900717

Haofeng Kou, Weijia Shang

{"title":"Parallelized feature extraction and acoustic model training","authors":"Haofeng Kou, Weijia Shang","doi":"10.1109/ICDSP.2014.6900717","DOIUrl":"https://doi.org/10.1109/ICDSP.2014.6900717","url":null,"abstract":"In this paper, we present our research on the parallelized speech recognition including both Mel-Frequency Cepstral Coefficient (MFCC) feature extraction [1] and Viterbi training for Hidden Markov Model (HMM) based acoustic model [2] on the Graphics Processing Units (GPU). Robust and accurate speech recognition systems can only be realized with adequately trained acoustic models derived from the effectively parsed features. For common languages, state-of-the-art systems are extracted and trained on many thousands of hours of speech data and even with large clusters of machines the entire extracting and training process can take weeks. To overcome this development bottleneck, we not only demonstrate that feature extraction and acoustic model training are suitable for GPUs, but also propose the optimized parallel implementation using highly parallel GPUs by combining the MFCC feature extraction along with Viterbi training for HMM acoustic model, illustrate its application concurrency characteristics, data working set sizes, and describe the optimizations required for effective throughput on GPU processors. We demonstrate that feature extraction and acoustic model training are well suited for GPUs. Using one GTX580 our approach is shown to be overall approximately 95x faster than a sequential CPU implementation at the same accuracy level, enabling feature extraction and acoustic model training to be performed at realtime.","PeriodicalId":301856,"journal":{"name":"2014 19th International Conference on Digital Signal Processing","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127879335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 5

Salient region detection for stereoscopic images 立体图像的显著区域检测

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900706

X. Fan, Zhi Liu, Guangling Sun

引用次数: 71

A new score normalization for text-independent speaker verification 一种新的文本无关说话人验证的评分归一化

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900743

H. Ning, Y. Zou, Xuyan Hu

引用次数: 1

Feature selection and channel optimization for biometric identification based on visual evoked potentials 基于视觉诱发电位的生物特征识别特征选择与通道优化

2014 19th International Conference on Digital Signal Processing Pub Date : 2014-09-18 DOI: 10.1109/ICDSP.2014.6900769

Yanru Bai, Zhiguo Zhang, Dong Ming

{"title":"Feature selection and channel optimization for biometric identification based on visual evoked potentials","authors":"Yanru Bai, Zhiguo Zhang, Dong Ming","doi":"10.1109/ICDSP.2014.6900769","DOIUrl":"https://doi.org/10.1109/ICDSP.2014.6900769","url":null,"abstract":"In recent years, biometric identification has received general concerns around the world, and become a frontal and hot topic in the information age. Among the internal biometric traits, electroencephalogram (EEG) signals have emerged as a prominent characteristic due to the high security, uniqueness and impossibility to steal or mimic. In this paper, individual difference of visual evoked potentials (VEPs) with cognition task were investigated, in addition, a feature selection and channel optimization strategy was developed for the VEPs based biometric identification system, where three different methods, including genetic algorithm (GA), Fisher discriminant ratio (FDR), and recursive feature elimination (RFE) were employed. In our experiments with 20 healthy subjects, the classification accuracy by support vector machine (SVM) reached up to 97.25% with AR model parameters, compared to 96.25% before optimization, and 32 channels of most discriminative were eventually selected from 64 channels with best performance. Results in this study revealed the feasibility of VEPs based EEG to be used for biometric identification. The proposed optimization algorithm was shown to have the ability to effectively improve the identification accuracy as well as simplifying the system. Further investigate may provide a novel idea for the individual difference analysis of EEG and for its practical design and optimization in the field of biometrics in the future.","PeriodicalId":301856,"journal":{"name":"2014 19th International Conference on Digital Signal Processing","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126579833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 16