GithubHelp home page GithubHelp logo

quanhua-guan / sketchretrieval Goto Github PK

View Code? Open in Web Editor NEW

This project forked from jamzywang/sketchretrieval

0.0 2.0 0.0 29.15 MB

A whole application of sketch retrieval written in matlab

MATLAB 99.58% Shell 0.28% Limbo 0.14%

sketchretrieval's Introduction

整个sketch retrieval的处理流程


1.离线处理(处理image)

(1)对所有image进行边缘提取

输入:所有原始图像

处理脚本(批处理)image_ edge_ detection.m

处理函数:AnisotropicDetector

输出:所有图像的边缘图像

备注:处理时间较长

(2)计算所有image的local feature

输入:(1)处理后得到的所有边缘图像

处理脚本(批处理)image_ local_ feature_extraction.m

处理函数: feature_ extraction_local.m

输出:所有图像的local feature

备注:处理时间较长,相比(1)时间要短

(3)计算图像的视觉词典codebook

输入:(2)得到的所有图像的local feature

处理脚本generate_local_feature_visual_word.m

输出:所有图像的local feature

备注:visual vocabulary用于量化local feature

(4)量化所有图像的local feature

输入:(2)得到的local feature,(3)得到的visual vocabulary

处理脚本(批处理)quantize_image_local_feature.m

处理函数: adaptive_weigthing_quantization.m

输出:所有图像经过量化后的local feature

备注

(5)计算所有image的分割情况

输入:(1)处理得到的边缘图像

处理脚本(批处理)image_divide.m

处理函数: divide_function.m

输出:每一个图像的分割情况

备注: 1)处理时间较快; 2)图像的分割情况可能用于匹配值计算(类似two-way-matching)

2.在线处理

2.1 处理sketch

(1)sketch预处理

输入:sketch

处理函数sketch_ processing.m

输出:sketch的边缘图像

备注:

(2)计算sketch的local feature

输入:(1)处理后的sketch

处理函数feature_ extraction_ local.m

输出:sketch的local feature

备注:

(3)量化sketch的local feature

输入:(2)处理后的local feature,离线得到的codebook

处理脚本adaptive_weigthing_quantization.m

输出:量化后的local feature

备注:

(4)计算sketch的分割情况

输入:(1)处理后的边缘图像

处理函数divide_function.m

输出:sketch的分割情况

备注:

2.2 计算global feature

2.2.1 根据sketch的分割情况计算global feature

(1) 根据sketch的分割情况计算sketch的global feature

输入:sketch的边缘图像,sketch的分割情况

处理函数feature_ extraction_ global

输出:sketch的global feature

备注:

(2) 根据sketch的分割情况计算image的global feature

输入:image的边缘图像,sketch的分割情况

处理函数feature_ extraction_ global

输出:image的global feature

(3)整理sketch和image的feature

对于sketch和image上的每一个兴趣点,现在已经可以得到每一个兴趣点的feature,每一个feature = [global feature, local feature]

2.2.2 根据image的分割情况计算global feature

(1) 根据image的分割情况计算sketch的global feature

输入:sketch的边缘图像,image的分割情况

处理函数feature_ extraction_ global

输出:sketch的global feature

备注:

(2) 根据image的分割情况计算image的global feature

输入:image的边缘图像,image的分割情况

处理函数feature_ extraction_ global

输出:image的global feature

(3)整理sketch和image的feature

对于sketch和image上的每一个兴趣点,现在已经可以得到每一个兴趣点的feature,每一个feature = [global feature, local feature]

###2.3 计算匹配值

2.3.1 计算sketch和所有图像的匹配值

输入:sketch,所有image

处理脚本retrieval.m

输出:最终的匹配值

备注:query_sketch.m中包含下面两部分

(1)计算sketch ——> image的匹配值

输入:sketch的feature,image的feature,sketch的分割情况

处理脚本calculate_ matching_ cost.m

输出:sketch ——> image的匹配值

备注

(2)计算image ——> sketch的匹配值

输入:sketch的feature,image的feature,image的分割情况

处理脚本calculate_ matching_ cost.m

输出:image ——> sketch的匹配值

备注

sketchretrieval's People

Contributors

jamzywang avatar

Watchers

James Cloos avatar Quanhua Guan avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.