Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

CDVS extraction process accelerating method based on GPGPU platform

An extraction process and platform technology, applied in the field of image processing, can solve problems such as the inability to meet performance requirements, and achieve the effect of reducing the time-consuming extraction process and CDV

Active Publication Date: 2016-11-09
PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF3 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] To sum up, the practical value of accelerating the MPEG-7CDVS extraction process is great, and the acceleration performance obtained by the existing acceleration methods is far from meeting the performance requirements of visual search applications based on this standard, but the powerful parallel computing capabilities of GPGPU Opens up new possibilities for better acceleration

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • CDVS extraction process accelerating method based on GPGPU platform
  • CDVS extraction process accelerating method based on GPGPU platform
  • CDVS extraction process accelerating method based on GPGPU platform

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0094] The specific implementation of this step is as follows:

[0095] 41) Allocate local memory for the feature points in the working group;

[0096] 42) During the execution of the kernel program, the access and update of the descriptor containing 128 floating-point numbers to the feature point occurs on the local memory allocated in the previous step;

[0097]43) After the operation logic of the descriptor calculation sub-stage is completed, the descriptor in the global memory is updated with the descriptor in the local memory.

[0098] In order to test the effectiveness of the present invention, the present invention uses the image set Graphics provided by the CDVS standardization organization for performance testing, and compares the time overhead of the CDVS initial extraction process and the accelerated extraction process. The input image size is fixed as a JPEG image of 640×480p, because the CDVS standard stipulates that all images whose length or width is greater th...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a CDVS extraction process accelerating method based on a GPGPU platform. For the two stages, feature detection and local descriptor computation, of CDVS extraction process, arithmetic logic and parallel strategy of inter-pixel parallel substages and inter-feature-point parallel substages are realized by means of an OpenCL universal parallel programming frame, and acceleration is achieved by means of the parallel computation capacity of a GPU. The method comprises the steps of dividing GPU and CPU computing tasks, reconstructing a scale space pyramid storage model of an image, assigning a parallel strategy to each substage on the GPU, and weakening memory access bottleneck by means of local memory. By the adoption of the technical scheme, CDVS extraction process can be accelerated, and extraction performance is improved remarkably.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to a visual search method, in particular to a CDVS (visual search compact descriptor) extraction process acceleration method based on a GPGPU platform. Background technique [0002] Visual search technology has developed rapidly in the past two decades, especially with the popularization of the Internet, various applications built around visual search have changed people's lifestyles to a large extent, such as image retrieval of search engines, monitoring applications object recognition, etc. Generally speaking, the starting point of visual search is to find a representative set of feature points in the image, and then perform related matching work based on this set of feature points. [0003] MPEG-7 documented in literature [1] (Duan L Y, Chandrasekhar V, Chen J, et al. Overview of the MPEG-CDVS Standard [J]. IEEE Transactions on Image Processing, 2015, 25(1): 323-332) The Vis...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/50G06F9/48
CPCG06F9/4887G06F9/5027G06F9/5038G06F9/5066G06T1/20G06F9/4881G06V10/454G06F9/50G06F9/3877G06F9/3885G06F17/10
Inventor 王荣刚张申王振宇高文
Owner PEKING UNIV SHENZHEN GRADUATE SCHOOL
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com