• <button id="nuo3m"><acronym id="nuo3m"></acronym></button>
    <dd id="nuo3m"></dd>
    <em id="nuo3m"></em>
    <button id="nuo3m"><object id="nuo3m"></object></button>
  • <s id="nuo3m"></s>
  • <th id="nuo3m"><track id="nuo3m"></track></th>
    首頁 > 科學研究 > 學術看板 > 正文

    圖像濾波中的半全局加權最小二乘

    供稿:    責任編輯:安果    時間:2018-05-18    閱讀:

    主講人:劉偉

    報告時間:20185219:00

    報告地點:基礎教學樓A303

    個人主頁:http://www.escience.cn/people/weiliusjtu/index.html

    報告題目:Semi-global Weighted Least Squares in Image Filtering

    (圖像濾波中的半全局加權最小二乘)


    報告摘要:

    圖像濾波中求解全局加權最小二乘(Weighted Least Squares, WLS)需要消耗大量的時間和內存。在本文中,我們提出了一個內存消耗和時間消耗都很小的替代算法,我們把它稱為半全局加權最小二乘(Semi-Global Weighted Least Squares, SG-WLS)。與加權最小而中求解一個大型的線性系統不同的是,我們提出迭代求解一系列一維加權最小二乘子系統。雖然每個子系統是一維的,但是由于我們提出了特殊的近鄰系統構造方法,這使得每個子系統能夠包含二維的近鄰信息。這個性質使得我們的半全局加權最小二乘能夠和原始的二維加權最小二乘有著相近的表現,但是我們的方法內存消耗和時間消耗更小。以往相關的算法只能夠處理4連接或者8連接近鄰系統,但是由于我們提出的快速求解算法,使得我們的半全局加權最小二乘能夠處理更泛化和更大的近鄰系統。這種泛化使得我們的算法能夠在一些應用中比4連接或8連接近鄰取得更好的結果。我們的半全局加權最小二乘比原始的加權最小二乘快~20倍。

    Abstract

    Solving the global method of Weighted Least Squares (WLS) model in image filtering is both time- and memory-consuming. In this paper, we present an alternative approximation in a time- and memory- efficient manner which is denoted as Semi-Global Weighed Least Squares (SG-WLS). Instead of solving a large linear system, we propose to iteratively solve a sequence of subsystems which are one-dimensional WLS models. Although each subsystem is one-dimensional, it can take two-dimensional neighborhood information into account due to the proposed special neighborhood construction. We show such a desirable property makes our SG-WLS achieve close performance to the original two-dimensional WLS model but with much less time and memory cost. While previous related methods mainly focus on the 4-connected/8-connected neighborhood system, our SG-WLS can handle a more general and larger neighborhood system thanks to the proposed fast solution. We show such a generalization can achieve better performance than the 4-connected/8-connected neighborhood system in some applications. Our SG-WLS is ~20 times faster than the WLS model. 


    報告人簡介:

    劉偉博士現為阿德萊德大學高級研究員。他博士就讀于上海交通大學大學。其主要研究興趣為圖像濾波及其在計算機視覺和圖形圖像學領域的應用。他在計算機視覺和機器學習領域國際會議(如ICCV,IJCAI)以及期刊(如TIP,TCSVT,TMM)上共發表10余篇論文。他于2017年被授予上海交通大學“學術之星”提名獎(上海交通大學研究生最高學術榮譽稱號)。

    Dr Wei Liu is a senior research fellow in the University of Adelaide. He had his doctoral research in Shanghai Jiao Tong University. His main research interests mainly focus on image filtering and its applications in computer vision and computational graphics. He has published more than 10 papers in the international conference/journals of computer vision and machine learning, including ICCV, IJCAI, TIP, TCSVT, TMM. He has been awarded the nomination of “Academic Star” of Shanghai Jiao Tong University which is the highest academic honor for graduate students in Shanghai Jiao Tong University.


    棋牌上下分