Adjacent Frame - Explore the Science & Experts | ideXlab

Scan Science and Technology

Contact Leading Edge Experts & Companies

Adjacent Frame

The Experts below are selected from a list of 2103 Experts worldwide ranked by ideXlab platform

Jun Zhou – 1st expert on this subject based on the ideXlab platform

  • FCCM – A High Throughput and Energy-Efficient Retina-Inspired Tone Mapping Processor
    2019 IEEE 27th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), 2019
    Co-Authors: Xiaoqiang Xiang, Yongjie Li, Jun Zhou

    Abstract:

    This paper presents a high throughput and energy-efficient retina inspired tone mapping processor. Several hardware design techniques have been proposed to achieve high throughput and high energy efficiency, including data partition based parallel processing with S-shape sliding, Adjacent Frame feature sharing, multi-layer convolution pipelining and convolution filter compression with zero skipping convolution. The proposed processor has been implemented on a Xilinx’s Virtex7 FPGA for demonstration. It is able to achieve a throughput of 189 Frames per second for 1024*768 RGB images with 819 mW. Compared with several state-of-the-art tone mapping processors, the proposed processor achieves higher throughput and energy efficiency. It is suitable for high-speed and energy-constrained video enhancement applications such as autonomous vehicle and drone monitoring.

  • A High Throughput and Energy-Efficient Retina-Inspired Tone Mapping Processor
    2019 IEEE 27th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM), 2019
    Co-Authors: Xiaoqiang Xiang, Yongjie Li, Jun Zhou

    Abstract:

    This paper presents a high throughput and energy-efficient retina inspired tone mapping processor. Several hardware design techniques have been proposed to achieve high throughput and high energy efficiency, including data partition based parallel processing with S-shape sliding, Adjacent Frame feature sharing, multi-layer convolution pipelining and convolution filter compression with zero skipping convolution. The proposed processor has been implemented on a Xilinx’s Virtex7 FPGA for demonstration. It is able to achieve a throughput of 189 Frames per second for 1024*768 RGB images with 819 mW. Compared with several state-of-the-art tone mapping processors, the proposed processor achieves higher throughput and energy efficiency. It is suitable for high-speed and energy-constrained video enhancement applications such as autonomous vehicle and drone monitoring.

Li Zeng – 2nd expert on this subject based on the ideXlab platform

  • SMC – Moving multi-object tracking algorithm based on wavelet clustering and Frame difference
    2009 IEEE International Conference on Systems Man and Cybernetics, 2009
    Co-Authors: Li Zeng

    Abstract:

    The paper presents an approach to motion objects tracking by combining parallel wavelet clustering with Frame difference, which satisfies certain requirements of rapid moving objects detection. By utilizing multi-resolution property of wavelet clustering analysis based on Adjacent Frame difference results, we can identify arbitrary shape moving objects at different degree of accuracy. Experiment results show that good accuracy of proposed algorithm can be obtained at speeds close to real-time. Applications in real world are also presented which further demonstrated the efficiency and effectiveness of the proposed method.

  • Motion objects detection based onwavelet clustering
    2009 2nd IEEE International Conference on Computer Science and Information Technology, 2009
    Co-Authors: Li Zeng, Wenjuan Wu

    Abstract:

    Motion objects detection is a process of segmenting the motion objects or regions from the rest of the sequent images. This paper presents an approach to motion objects detection by applying wavelet clustering with Frame difference, which satisfies some requirements of rapid motion objects detection. By analyzing Adjacent Frame difference results and utilizing multi-resolution property of wavelet clustering analysis, we can identify arbitrary shape multiply motion objects. Experiment results show that good accuracy of proposed algorithm was obtained.

  • Moving multi-object tracking algorithm based on wavelet clustering and Frame difference
    2009 IEEE International Conference on Systems Man and Cybernetics, 2009
    Co-Authors: Li Zeng, Lida Xu

    Abstract:

    The paper presents an approach to motion objects tracking by combining parallel wavelet clustering with Frame difference, which satisfies certain requirements of rapid moving objects detection. By utilizing multi-resolution property of wavelet clustering analysis based on Adjacent Frame difference results, we can identify arbitrary shape moving objects at different degree of accuracy. Experiment results show that good accuracy of proposed algorithm can be obtained at speeds close to real-time. Applications in real world are also presented which further demonstrated the efficiency and effectiveness of the proposed method.

Xuqiao Zhang – 3rd expert on this subject based on the ideXlab platform

  • behavior of steel plate shear walls with pre compression from Adjacent Frame columns
    Thin-walled Structures, 2014
    Co-Authors: Xuqiao Zhang

    Abstract:

    Abstract This paper investigates the behavior of steel plate shear walls (SPSWs) with pre-compression from Adjacent Frame columns which is produced in the construction process. Firstly, some parameters used in analytical finite element models, such as the stiffness of Frame beams and columns and the magnitude of the loads are discussed. Then, numbers of numerical examples are analyzed and show that the influence of pre-compression varies with the dimension of SPSWs. Also, the distribution and transferring of axial forces between Frame columns and SPSWs during loading are discussed. Finally, a reduction coefficient of shear-carrying capacity of SPSW due to pre-compression is proposed.