Overview and Classification of Parallel Architectures for Video Signal Processors

Da-nian GONG,Yun HE,Zhi-gang CAO

DOI: https://doi.org/10.3321/j.issn:0372-2112.2000.07.025

2000-01-01

Abstract:A new classification of video signal processors based on parallelism is first pr esented in this paper. Then an overview of parallel architectures of video signal processors is given and the future development of parallel architectures is poi nted out.

What problem does this paper attempt to address?

Scalable SVM Processor and Its Application to Nonlinear Channel Equalization

Kuikang Cao,Haibin Shen

DOI: https://doi.org/10.1109/paccs.2009.157

2009-01-01

Abstract:Support vector machines (SVMs) are powerful, state-of-the-art machine learning tools. With the aim to integrate SVM training capability into embedded systems while being able to meet area and performance constraints, a parallel and scalable digital architecture for training SVMs on-line is proposed and implemented on a field-programmable gate array (FPGA). Experiments show that the proposed SVM processor can solve the channel equalization problem effectively with fixed-point arithmetic and exhibits good scalability with respect to the number of cache update units (CUUs). This architecture is particularly suitable to be applied in embedded environments, as designers can easily trade off between area and performance to meet the constraints.
Scalable Parallel Computers for Real-Time Signal Processing

H Kai,ZW Xu

DOI: https://doi.org/10.1109/79.526898

IF: 15.204

1996-01-01

IEEE Signal Processing Magazine

Abstract:We assess the state-of-the-art technology in massively parallel processors (MPPs) and their variations in different architectural platforms. Architectural and programming issues are identified in using MPPs for time-critical applications such as adaptive radar signal processing. We review the enabling technologies. These include high-performance CPU chips and system interconnects, distributed memo...
Video Stream Processing on a High Performance Reconfigurable Architecture

Tao Li,Zhentao Liu

DOI: https://doi.org/10.1109/dicta.2011.72

2011-01-01

Abstract:A dynamically reconfigurable architecture VSP-DR1 for video processing is proposed in this paper. This architecture is derived from years of engineering video ASICs and is ideally suited to real-time video streaming. The various cells of the architecture are optimized for video post processing. The architecture, along with its video application, can be used in many real-time video applications.
A Parallel Serial Filtering Mixed Advanced ID Interpolation Architecture for AVS

Kai Luo,Dongxiao Li,Yu Dai,Zheng Zhu,Ming Zhang

DOI: https://doi.org/10.1109/icasic.2007.4415742

2007-01-01

Abstract:This paper presents an advanced ID interpolation architecture for AVS video coding standard. Previous approaches have demerit of low data throughput. The proposed design parallel filters the horizontal data and serial filters the vertical data, thus greatly improve the throughput. Experimental result shows it outperforms conventional designs in processing cycle and chip area. The prototype design costs 26.2 K gates when synthesized at 108 MHz using 0.18 mum CMOS technology.
Novel many-core architecture design for real-time image processing

LIUZhentao,LI Tao,HUANG Hucai,HAN Jungang,SHEN Xubang

DOI: https://doi.org/10.3969/j.issn.1001-2400.2015.02.016

2015-01-01

Journal of Xidian University

Abstract:Based on the data-flow model and hardware reconfigurable technology, a polymorphic reconfigurable many-core processor architecture is presented for image processing. It is a scalable hierarchically organized parallel architecture,which is capable of supporting a dynamic mixture of multiple parallel computing models,and overcomes the inefficiency of traditional data-flow implementation by using distributed shared memory and neighbor interconnect architecture with hardware handshaking.From the beginning of the architecture design,based on the VC++,the integrated simulation platform (ISE) is developed for verifying the architecture and the performance of the instruction set.In addition,we also implement the proposed architecture on the FPGA.Experimental results show that the architecture can be used in many image processing applications,and achieve the throughput close to that of the ASIC and the performance better than that of the GPU.
Overview of Processor Architecture for Media Processing

HE Jing,SUN Yihe

DOI: https://doi.org/10.3969/j.issn.1000-3428.2007.09.081

2007-01-01

Abstract:Efficient processing of multimedia application is becoming one of the main bottlenecks in media application field,and many architectures have been proposed for multimedia processing.This paper classifies the contemporary processors into three groups based on programmability and evolution process:dedicated media processors,general-purpose processors with media ISA extension and multi-core media processors.An overview of media processor architectures is presented with chip examples,and the future development of media processor architectures is pointed out.
Survey of Parallel Acceleration Algorithms of Video Coding on Multi-core Processor

Shiqiang Yang

2009-01-01

Abstract:The high complexity of video processing algorithm demands high computing capability of processor,and multi-core processor provides a new platform for multimedia processing.The characteristics of video coding are analyzed,the methods of accelerating video coding are summarized,and the schemes and examples of accelerating video coding on multi-core processor are introduced by classifying multi-core into symmetric,asymmetric and hybrid multi-core processor.The difficulties of designing video coding algorithm on multi-core processor are illuminated,and finally the future works are demonstrated.
Design and Research of Parallel Image Information Processing Platform Based on VPX: In This paper, The network topology of the VPX bus standard is studied, and the author proposes and implements a more efficient way.

Gen Wang,Qing-Shan Liu,Hai-Yang Zhou,Na Wan,Lin Ren,Ying-Xue Li,Jie Zhou,Guang Yang

DOI: https://doi.org/10.1145/3532213.3532322

2022-01-01

Abstract:Digital signal processing theory maturing, increasing the amount of signal processing algorithms, data processing algorithms become more complex, resulting in the past restricted the kind of hardware platform program performance, high flexibility, high real-time and high data throughput application. Therefore, a large-scale parallel platform with high real-time, high precision and high data throughput is necessary. This paper presents a highly flexible, large amount of data and high-performance parallel network processing platform based on VPX standard. The designer developed a parallel image information processing platform based on this architecture. This solution aims to dynamically reconstruct different message channels through the VPX switch module to reduce the complexity and cost of the platform.
Parallelization Design of Irregular Algorithms of Video Processing on GPUs.

Huayou Su,Jun Chai,Mei Wen,Ju Ren,Chunyuan Zhang

DOI: https://doi.org/10.1109/icme.2012.147

2012-01-01

Abstract:In this paper, we present the parallelization design consideration for irregular algorithms of video processing on GPUs. Enrich parallelism can be exploited by scheduling the processing order or making a tradeoff between performance and parallelism for irregular algorithms (such as CAVLC and deblocking filter). We implement a component-oriented CAVLC encoder and a direction-oriented deblocking filter on GPUs. The experiment results show that, compared with the implementation on CPU, the optimized parallel methods achieve high performance in term of speedup ratio from 63 to 44, relatively for deblocking filter and CAVLC. It shows that the rich parallelism is one of the most important factors to gain high performance for irregular algorithms based on GPUs. In addition, it seems that for some irregular kernels, the number of SM of GPU is more important to the performance than the computation capability.
Parallel System Architecture of Multi-Dsp Interconnected by FPGA

YAN Lu-xin,ZHANG Tian-xu,ZOU sheng,ZHONG sheng

DOI: https://doi.org/10.3321/j.issn:1001-506x.2005.10.026

2005-01-01

Abstract:To meet real-time image processing demands,the paper analyses the architectures of common parallel systems and presents a DSP parallel system architecture based on FPGA.This architecture is structured with high performance DSPs interconnected by PGA.Within FPGA a FIFO interconnection network and the specific data communication protocol are implemented,which interconnect 3 DSPs(TMS320C6414) effectively.The system architecture is reconfigurable and scalable.The measured performances in the prototype with the proposed parallel architecture show high data transfer rate as well as low latency between DSPs and meet parallel real-time processing demands.
Heterogeneous Multi-core Systems for Video Processing

Bingbing Xia,Fei Qiao,Huazhong Yang,Hui Wang

DOI: https://doi.org/10.2174/1874476111003030200

2010-01-01

Recent Patents on Electrical Engineering

Abstract:Multi-core system is the future of the embedded processor design for its power efficiency, multi-thread parallelization and flexibility. Meanwhile, with the improvement of the video processing algorithm, the processing capability requirement is also on the increase to meet such high processing capability requirement, the embedded multicore processor is developed as an appropriate choice. To better improve the performance of such systems, heterogeneous multi-core is preferred than homogeneous multi-core. For the design of such heterogeneous multi-core systems for video processing, the first and the most important step is to implement the system-level design and this paper gives an overview of such system-level design methodologies and some new methods for both cycle-approximate transaction-level modeling and cycle-accurate transaction-level modeling are given in detail. In addition, the system-level design framework for such systems is proposed and a case study of a heterogeneous multi-core h.264 video decoder shows that it can be used to design such systems effectively. The recent patents used for such designs are also listed accordingly. © 2010 Bentham Science Publishers Ltd.
A Parallel Reconfigurable Architecture for Real-Time Stereo Vision

Lei Chen,Yunde Jia

DOI: https://doi.org/10.1109/icess.2009.18

2009-01-01

Abstract:In this paper, a parallel reconfigurable architecture is proposed for real-time stereo vision computation. The architecture is divided into four components., input port, output port, memory and processor. We use task partition methods to achieve the maximum parallel and full pipeline processing of the algorithm implementation. We also adopt memory management to decrease the latency of memory access time and accelerate the processing speed. Data bandwidth control is employed to reduce the hardware resource consumption while maintaining precision demand of computation. Based on the proposed architecture and design method, we have developed a miniature stereo vision machine (MSVM33) to generate high-resolution dense disparity maps at the video rate for real-time applications.
A Parallel Memory Architecture for Video Coding

Jian-ying Peng,Xiao-lang Yan,De-xian Li,Li-zhong Chen

DOI: https://doi.org/10.1631/jzus.a0820052

2008-01-01

Abstract:To efficiently exploit the performance of single instruction multiple data (SIMD) architectures for video coding, a parallel memory architecture with power-of-two memory modules is proposed. It employs two novel skewing schemes to provide conflict-free access to adjacent elements (8-bit and 16-bit data types) or with power-of-two intervals in both horizontal and vertical directions, which were not possible in previous parallel memory architectures. Area consumptions and delay estimations are given respectively with 4, 8 and 16 memory modules. Under a 0.18-μm CMOS technology, the synthesis results show that the proposed system can achieve 230 MHz clock frequency with 16 memory modules at the cost of 19k gates when read and write latencies are 3 and 2 clock cycles, respectively. We implement the proposed parallel memory architecture on a video signal processor (VSP). The results show that VSP enhanced with the proposed architecture achieves 1.28× speedups for H.264 real-time decoding.
Simd Massively Parallel Processing System For Real-Time Image Processing

Xiaochu Chen,Ming Zhang,Qingdong Yao,Jilin Liu,Hong Ye,Song Wu,Dongxiao Li,Yong Zhang,Lei Ding,Zhongyang Yao,Weijian Yang,Qiaohai Pan

DOI: https://doi.org/10.1117/12.323605

1998-01-01

Abstract:This paper will describe the embedded SIMD massively parallel processor that we have developed for real-time image processing applications, such as real-time small target detection and tracking and video processing. The processor array is based on SIMD chip BAP-128 designed by our own, and uses high performance DSP TMS320C31, which can effectively perform serial and floating point calculations, as the host of the SIMD processor array. As a result, the system is able to perform a variety of image processing tasks in real-time. Furthermore, the processor will be connected with a MIMD parallel processor to construct a heterogeneously parallel processor for more complex realtime ATR (Automatic Target Recognition) and computer vision applications.
Design of H.264 Parallel Decoding for Ultra-HD Video

Xiansu Gao,Yonglin Xue

DOI: https://doi.org/10.1109/ICECC.2012.435

2012-01-01

Abstract:Video decode application has been with high demands on the hardware. Especially for the 2160p ultra-HD video, we cannot achieve real time decoding now. Along with the popularity of multi processors using, the decoder application in parallel on multi processors environment to improve decoding performance has been an effective way. This paper analyzes the dependencies between data blocks, proposes a resolution of images partition method on macro blocks and designs a multi-threaded parallel decoding process. Test results show that in the macro block level of H.264 parallelization run in multi processor environment can give full play performance.
Parallel Interactive Ray-Space Rendering Approach And Implementation

He Wang,GuoZhong Wang,GuoWei Teng,Zhaoyang Zhang

DOI: https://doi.org/10.1109/SOPO.2009.5230176

2009-01-01

Abstract:A real time interactive ray-space rendering approach is proposed. Multiview video(MVV) becomes more and more important in the field of image processing, and FTV as an application of MVV based on ray-space is required to be interactive and real time. Both of the state-of-the-art rendering techniques ray-space and Lightfield have to handle enormous data, thus the complexity of rendering multiviews in real time is very high. The common approaches based on sequential processing or multi-core processing just can render 4 views in real time at most. The proposal can render 16 views which can be controlled by the view control interactive signals in real time. The parallel architecture divided the 16 views into 4 view groups (GOP). The GOPs are being processed simultaneously, thus the rendering time is decreased to 1.08ms/frame. The simulation shows that the proposal has the same rendering quality as the rendering on software, we implement the approach on FPGA platform, and also test the FPGA design on Modelsim. The implementation results show that the logic resource cost on FPGA platform is only 139 slices.
A high speed multi-level-parallel array processor for vision chips

Cong Shi,Jie Yang,NanJian Wu,ZhiHua Wang

DOI: https://doi.org/10.1007/s11432-013-4969-4

2014-01-01

Science China Information Sciences

Abstract:This paper proposes a high speed multi-level-parallel array processor for programmable vision chips. This processor includes 2-D pixel-parallel processing element (PE) array and 1-D row-parallel row processor (RP) array. The two arrays both operate in a single-instruction multiple-data (SIMD) fashion and share a common instruction decoder. The sizes of the arrays are scalable according to dedicated applications. In PE array, each PE can communicate not only with its nearest neighbor PEs, but also with the next near neighbor PEs in diagonal directions. This connection can help to speed up local operations in low-level image processing. On the other hand, global operations in mid-level processing are accelerated by the skipping chain and binary boosters in RP array. The array processor was implemented on an FPGA device, and was successfully tested for various algorithms, including real-time face detection based on PPED algorithm. The results show that the image processing speed of proposed processor is much higher than that of the state-of-the-arts digital vision chips.
Novel Multicore Video Processing Architecture

CHEN Jianwen,YU Rong,MEI Shunliang

DOI: https://doi.org/10.3321/j.issn:1000-0054.2008.01.019

2008-01-01

Abstract:Multicore architectures provide high computational capacity and high throughput, so they are very useful for video processing applications. However, the increasing designs complexity and cost of the multicore system make it very difficult to use. Video processing applications need a generic, flexible video processing multicore platform for a wide variety of video standards. This paper presents a flexible heterogeneous multicore architecture system on video applications. An MPEG4 simple profile decoder was implemented on the multicore architecture. The performance shows that the generic, flexible media-processing design platform architecture can efficiently enhance video processing applications.
Design of Reconfigurable Architectures for Multi-Standards Video Decoder

Sisi Tan,Fei Qiao,Hua Z. Yang,Hui Wang

DOI: https://doi.org/10.2174/1874476111104010063

2011-01-01

Recent Patents on Electrical Engineeringe

Abstract:Digital video decoding is a typical data-intensive video processing application. Currently, high throughput and real time processing are the fundamental demands of performance for video processing systems. With the development of various video standards, multi-standard applications have become another key feature. Thus, the high performance, low area cost and low power consumption make up the most important design targets when realizing video processing chips. Reconfigurable hardware architectures have been reviewed in this paper, which could meet the above requirements for multi-standard video decoders. Traditional reconfiguration methods usually decrease area cost by reconfiguring the interconnections among function units. A relatively new approach is dynamic reconfiguration, which could reconfigure hardware resources at runtime. Recently, a new method called Reconfigurable Video Coding (RVC) has come up, which is a library-based method to design a reconfigurable system for multi-standard video decoders. An overview of the methodologies for reconfigurable video processing system is shown in the paper, as well as case studies, which demonstrate the effectiveness of the design flow. The article presents some promising patents on design of reconfigurable architectures for multi-standards video decoder. Keywords: Reconfigurable hardware architecture, video decoder, multi-standard, dynamic reconfiguration, reconfigurable interconnection, reconfigurable video coding (RVC), real time processing, compressed video streams, real-time restriction, FPGA, ASIC implementation, System-on-Chip (SoC), granularities, runtime, coarse granularity, fine granularity, Multi-standard video encoding/decoding, bitstream, reconfigured modules, MPEG series, ISO, International Telecommunication Union (ITU), AVS, Audio Video coding Standard, H.26x series, Joint Video Team (JVT), high-density storage, resolution mobility, De-blocking Filter, interconnection reconfiguration, logic array reconfiguration, abstract system level, Library-based flow, dataflow graphs, CDDF, BDF, BDDF, token rate, synchronous dataflow, Dynamic Graph Topology (DGT), High Level Reconfiguration (HLR), Validation (VAL), Low Level Reconfiguration (LLR), reconfigurable IP cores, FU library, interconnection topology, syntax, semantics, MPEG-2/H.264 Dual-Standard Video Decoder, SystemC language, dual-standard decoder, Register Transfer Level (RTL), Hardware Description Language, behavior-level model, reference data reading, interpolation sub-modules, picture reconstruction, FIFO, DDR interface, header parser, inverse quantization, inverse transform, motion vector decoding and controller, QCIP video sequence decoding, RefRead
A study on parallel processing for high-definition video

Feng Wei,Yuan Jie

DOI: https://doi.org/10.13232/j.cnki.jnju.2012.01.004

2012-01-01

Abstract:High-definition(HD) video processing is always in company with high bandwidth and enormous calculated amount.In order to reduce bandwidth,new video compression algorithms with high compression ratio continue to emerge.However,improvement in compression ratio causes increasing complexity of algorithm.Therefore,how to improve computational efficiency has become a hot research topic.To achieve this goal,there are many methods,among which parallel computing is a very important one.It has lower cost to improve system performance.Because of the importance of parallel computing,many processors have their own structures to achieve parallelism,and parallel processing capacity has become an important measurement of the performance of a processor.In this study,we make full use of a digital signal processor(DSP) with parallel computing capabilities to construct a high-definition video processing system.We accomplished both hardware and software development of the system.In the hardware structure,a high definition video capture integrated circuit(IC) is used,which is able to output HD Bayer format images directly.The DSP we use could deal with Bayer format itself,therefore the capture IC could connect to the DSP video port directly,without the need of additional transformation.This hardware structure does not only improve the system's integration,but also reduce the development time and costs.We developed a new algorithm which is based on absolute moment block truncation coding to compress Bayer images.The algorithm makes fully consideration of the complexity and the compression ratio.It is a balanced solution.It is able to reduce the signal bandwidth,so that the output video stream is suitable for network transmission.What's more,this algorithm is computational affordable.The Bayer compression algorithm can reduce the bandwidth which is essential for network transmission.In order to improve stability and maintainability of the software program,the program uses a framework which is specially developed for video processing program.Thus,we don't need to consider the underlying drivers and the reliability of collaborations between tasks.This system can be used for traffic surveillance system and other fields,and it can both reduce the cost of monitoring systems and improve reliability.

Overview and Classification of Parallel Architectures for Video Signal Processors

Scalable SVM Processor and Its Application to Nonlinear Channel Equalization

Scalable Parallel Computers for Real-Time Signal Processing

Video Stream Processing on a High Performance Reconfigurable Architecture

A Parallel Serial Filtering Mixed Advanced ID Interpolation Architecture for AVS

Novel many-core architecture design for real-time image processing

Overview of Processor Architecture for Media Processing

Survey of Parallel Acceleration Algorithms of Video Coding on Multi-core Processor

Design and Research of Parallel Image Information Processing Platform Based on VPX: In This paper, The network topology of the VPX bus standard is studied, and the author proposes and implements a more efficient way.

Parallelization Design of Irregular Algorithms of Video Processing on GPUs.

Parallel System Architecture of Multi-Dsp Interconnected by FPGA

Heterogeneous Multi-core Systems for Video Processing

A Parallel Reconfigurable Architecture for Real-Time Stereo Vision

A Parallel Memory Architecture for Video Coding

Simd Massively Parallel Processing System For Real-Time Image Processing

Design of H.264 Parallel Decoding for Ultra-HD Video

Parallel Interactive Ray-Space Rendering Approach And Implementation

A high speed multi-level-parallel array processor for vision chips

Novel Multicore Video Processing Architecture

Design of Reconfigurable Architectures for Multi-Standards Video Decoder

A study on parallel processing for high-definition video