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

High-dimensional-data-oriented vocabulary tree building method based on heterogeneous platform

A heterogeneous platform and construction method technology, which is applied in the fields of electronic digital data processing, program startup/switching, program control design, etc., and can solve the problems of inconvenient application of image search at the time of tree building

Inactive Publication Date: 2016-05-11
FUDAN UNIV
View PDF10 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Once the image database is changed, it is necessary to re-build the new image data set. Such a long time to build the tree brings great inconvenience to the image search application.

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
  • High-dimensional-data-oriented vocabulary tree building method based on heterogeneous platform
  • High-dimensional-data-oriented vocabulary tree building method based on heterogeneous platform
  • High-dimensional-data-oriented vocabulary tree building method based on heterogeneous platform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The specific implementation manner of the present invention will be described below in conjunction with the accompanying drawings of the present invention. The invention mainly utilizes the powerful parallel computing capability and programmability of the graphics processor in the heterogeneous platform to realize the acceleration of the high-dimensional vocabulary tree construction process. The specific process is as Figure 5 As shown, the specific steps are as follows:

[0037] (1) Read the feature point data from the file, create the initial node, put it into the task team, and hand it over to the tree building program to run.

[0038] (2) The main thread obtains a node from the head of the queue, selects the implementation method of generating a new node according to the size of the node, and coordinates the workload between GPUs:

[0039] (a) Large-scale node: The node cannot be loaded into the GPU memory as a whole. The node is divided into data blocks that can...

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 belongs to the technical field of a parallel processing unit, and particularly relates to a high-dimensional-data-oriented vocabulary tree building method based on a heterogeneous platform. According to the method, the strong parallel calculation capability and the programmability of a graphics processing unit arranged on a heterogeneous processing platform (a mixed architecture of a general purpose processor and a GPGPU (General Purpose Graphics Processing Unit)) are used for accelerating the speed of the building process of a high-dimensional vocabulary tree. The characteristic of high concurrency of the graphics processing unit is used for accelerating the speed of the core process of a high-dimensional vocabulary tree algorithm; the characteristics of the high-dimensional vocabulary tree algorithm and the memory access mode of the graphics processing unit are used for optimizing the memory accessing process of the algorithm; and a coordination strategy of a host and the graphics processing unit in the operation process of the high-dimensional vocabulary tree algorithm is designed. The method has the advantage that the building speed of the high-dimensional-data-oriented vocabulary tree can be effectively accelerated.

Description

technical field [0001] The invention belongs to the technical field of parallel processors, and in particular relates to a high-dimensional data-oriented vocabulary tree (VocabularyTree) construction method based on heterogeneous platform processing capabilities. Background technique [0002] With the advent of the era of big data, massive data has penetrated into every industry in society, and the processing of massive data has become a hot topic of discussion. In the field of image retrieval, a large number of data pictures brings unprecedented difficulties to retrieval. According to statistics, Facebook uploads 350 million pictures and 7PB of data every day. For such a huge amount of data, how to quickly and effectively retrieve it to provide a better user experience has become a huge challenge. The proposal of high-dimensional VocabularyTree (vocabulary tree) effectively improves the performance of image recognition. The algorithm first uses the local feature extraction...

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
IPC IPC(8): G06F9/48
CPCG06F9/4881
Inventor 张为华季晓枫余时强
Owner FUDAN UNIV
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