Article View: csd.grad
Article #143[CFAR Seminars] REMINDER: CVL Seminar - Mon. Nov. 24 at 11:00 am in Room 2120. Speaker: Allen Yang from UC Berkeley
From: "Janice Perrone"
Date: Mon, 24 Nov 2008 15:00
Date: Mon, 24 Nov 2008 15:00
390 lines
18006 bytes
18006 bytes
This is a multi-part message in MIME format.
--===============0642952306==
Content-class: urn:content-classes:message
Content-Type: multipart/alternative;
boundary="----_=_NextPart_001_01C94E45.6B9DE14C"
This is a multi-part message in MIME format.
------_=_NextPart_001_01C94E45.6B9DE14C
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
C V L S E M I N A R
=20
Monday, November 24, 11:00 AM
=20
Computer Vision Laboratory
Center for Automation Research
A.V. Williams Bldg., Room 2120
University of Maryland
=20
=20
High-Dimensional Multi-Model Estimation -- Its Algebra, Statistics,=20
and Sparse Representation
=20
Allen Y. Yang
Research Engineer
UC Berkeley
=20
ABSTRACT
=20
Recent advances in information technologies have led to unprecedented
large amounts of high-dimensional data from many emerging applications.
The need for more advanced techniques to analyze such complex data calls
for shifting research paradigms. In this talk, I will overview and
highlight several results in the area of estimation of mixture models in
high-dimensional data spaces. Applications include motion segmentation,
image segmentation, face recognition, and human action categorization.
Through this talk, I intend to emphasize the confluence of algebra and
statistics that may lead to more advanced solutions in analyzing complex
singular data structures such as mixture linear subspaces and nonlinear
manifolds.
=20
In the first part of the talk, I will start by reviewing a solution to
simultaneously segment and estimate mixture subspace models --
Generalized Principal Component Analysis (GPCA). In contrast to
traditional statistical methods, GPCA is focused on recovering a set of
vanishing polynomials that globally determines mixture subspaces as its
zero set. I will introduce a new algebro-geometric framework along the
same approach to simultaneously segment mixture quadratic manifolds. The
new solution is also robust to moderate data noise and outliers. The
second part will be focused on classification of mixture subspace
models, where the prior information of mixture subspaces is provided
through training examples. Inspired by compressive sensing theory, the
recognition problem can be reformulated via a sparse representation.
Furthermore, efficient solutions exist to recover such sparse
representation using fast L-1 minimization. Finally, I will discuss
several open problems in the emerging field of distributed sensor
perception.
=20
Bio:
Allen Y. Yang is a postdoctoral researcher in the department of EECS at
UC Berkeley. His primary research is in pattern analysis of geometric
and statistical models in high-dimensional data spaces, and applications
in motion segmentation, image segmentation, face recognition, and
distributed sensor perception. He received a BEng in Computer Science
from the University of Science and Technology of China (USTC) in 2001.
He received an MS in Electrical Engineering in 2003, an MS in
Mathematics in 2005, and a PhD in Electrical and Computer Engineering in
2006, all from the University of Illinois at Urbana-Champaign (UIUC). He
has published five journal papers and more than 10 conference papers.
=20
For more information, contact Janice M. Perrone at (301) 405-1736 or
janice@umiacs.umd.edu
=20
------_=_NextPart_001_01C94E45.6B9DE14C
Content-Type: text/html;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" =
xmlns=3D"http://www.w3.org/TR/REC-html40">
<head>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<meta name=3DGenerator content=3D"Microsoft Word 12 (filtered medium)">
<style>
<!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Times;
panose-1:2 2 6 3 5 4 5 2 3 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:Times;}
p.MsoTitle, li.MsoTitle, div.MsoTitle
{mso-style-priority:10;
mso-style-link:"Title Char";
margin:0in;
margin-bottom:.0001pt;
text-align:center;
font-size:36.0pt;
font-family:Times;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.TitleChar
{mso-style-name:"Title Char";
mso-style-priority:10;
mso-style-link:Title;
font-family:Times;}
span.EmailStyle19
{mso-style-type:personal;
font-family:"Calibri","sans-serif";
color:windowtext;}
span.EmailStyle20
{mso-style-type:personal;
font-family:"Calibri","sans-serif";
color:#1F497D;}
span.EmailStyle21
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page Section1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.Section1
{page:Section1;}
-->
</style>
<!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=3DEN-US link=3Dblue vlink=3Dpurple>
<div class=3DSection1>
<p class=3DMsoTitle>C V L S E M I N A R<span =
style=3D'font-size:10.0pt'><o:p></o:p></span></p>
<p class=3DMsoTitle><span =
style=3D'font-size:10.0pt'><o:p> </o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><b><span
style=3D'font-size:14.0pt'>Monday, November 24, 11:00 =
AM<o:p></o:p></span></b></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><span
style=3D'font-size:20.0pt'><o:p> </o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'>Computer =
Vision
Laboratory<o:p></o:p></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'>Center =
for Automation
Research<o:p></o:p></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><b>A.V. =
Williams
Bldg., Room 2120<o:p></o:p></b></p>
<p class=3DMsoNormal align=3Dcenter =
style=3D'text-align:center'>University of
Maryland<o:p></o:p></p>
<p class=3DMsoTitle><span =
style=3D'font-size:10.0pt'><o:p> </o:p></span></p>
<p class=3DMsoTitle><span =
style=3D'font-size:10.0pt'><o:p> </o:p></span></p>
<p class=3DMsoNormal align=3Dcenter =
style=3D'text-align:center;text-autospace:none'><b><span
style=3D'font-family:"Times New Roman","serif"'>High-Dimensional =
Multi-Model Estimation
-- Its Algebra, Statistics, <o:p></o:p></span></b></p>
<p class=3DMsoNormal align=3Dcenter =
style=3D'text-align:center;text-autospace:none'><b><span
style=3D'font-family:"Times New Roman","serif"'>and Sparse =
Representation<o:p></o:p></span></b></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><b><span
style=3D'font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></b></p>
<p class=3DMsoNormal align=3Dcenter =
style=3D'text-align:center;text-autospace:none'><b><span
style=3D'font-family:"Times New Roman","serif"'>Allen Y. =
Yang<o:p></o:p></span></b></p>
<p class=3DMsoNormal align=3Dcenter =
style=3D'text-align:center;text-autospace:none'>Research
Engineer<span style=3D'font-size:10.0pt;font-family:"Times New =
Roman","serif"'><o:p></o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'>UC =
Berkeley<span
style=3D'font-family:"Times New Roman","serif"'><o:p></o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><span
style=3D'font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><span
style=3D'font-size:10.0pt;font-family:"Times New =
Roman","serif"'>ABSTRACT<o:p></o:p></span></p>
<p class=3DMsoNormal style=3D'text-align:justify'><span =
style=3D'font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New Roman","serif"'>Recent =
advances
in information technologies have led to unprecedented large amounts of =
high-dimensional
data from many emerging applications. The need for more advanced =
techniques to
analyze such complex data calls for shifting research paradigms. In this =
talk,
I will overview and highlight several results in the area of estimation =
of mixture
models in high-dimensional data spaces. Applications include motion
segmentation, image segmentation, face recognition, and human action
categorization. Through this talk, I intend to emphasize the confluence =
of
algebra and statistics that may lead to more advanced solutions in =
analyzing
complex singular data structures such as mixture linear subspaces and =
nonlinear
manifolds.<o:p></o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New Roman","serif"'>In the =
first
part of the talk, I will start by reviewing a solution to simultaneously
segment and estimate mixture subspace models -- Generalized Principal =
Component
Analysis (GPCA). In contrast to traditional statistical methods, GPCA is
focused on recovering a set of vanishing polynomials that globally =
determines
mixture subspaces as its zero set. I will introduce a new =
algebro-geometric
framework along the same approach to simultaneously segment mixture =
quadratic
manifolds. The new solution is also robust to moderate data noise and =
outliers.
The second part will be focused on classification of mixture subspace =
models,
where the prior information of mixture subspaces is provided through =
training
examples. Inspired by compressive sensing theory, the recognition =
problem can
be reformulated via a sparse representation. Furthermore, efficient =
solutions
exist to recover such sparse representation using fast L-1 minimization.
Finally, I will discuss several open problems in the emerging field of
distributed sensor perception.<o:p></o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New =
Roman","serif"'>Bio:<o:p></o:p></span></p>
<p class=3DMsoNormal =
style=3D'text-align:justify;text-autospace:none'><span
style=3D'font-size:10.0pt;font-family:"Times New Roman","serif"'>Allen =
Y. Yang is
a postdoctoral researcher in the department of EECS at UC Berkeley. His =
primary
research is in pattern analysis of geometric and statistical models in
high-dimensional data spaces, and applications in motion segmentation, =
image
segmentation, face recognition, and distributed sensor perception. He =
received
a BEng in Computer Science from the University of Science and Technology =
of
China (USTC) in 2001. He received an MS in Electrical Engineering in =
2003, an
MS in Mathematics in 2005, and a PhD in Electrical and Computer =
Engineering in
2006, all from the University of Illinois at Urbana-Champaign (UIUC). He =
has
published five journal papers and more than 10 conference =
papers.<o:p></o:p></span></p>
<p class=3DMsoNormal><span style=3D'font-family:"Times New =
Roman","serif"'><o:p> </o:p></span></p>
<p class=3DMsoNormal align=3Dcenter style=3D'text-align:center'><span
style=3D'font-size:10.0pt'>For more information, contact Janice M. =
Perrone at
(301) 405-1736 or janice@umiacs.umd.edu</span><span =
style=3D'font-size:26.0pt'><o:p></o:p></span></p>
<p class=3DMsoNormal><span =
style=3D'font-size:11.0pt;font-family:"Calibri","sans-serif"'><o:p> =
</o:p></span></p>
</div>
</body>
</html>
------_=_NextPart_001_01C94E45.6B9DE14C--
--===============0642952306==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
_______________________________________________
cfar-Seminar mailing list
cfar-Seminar@mailman.umiacs.umd.edu
http://mailman.umiacs.umd.edu/mailman/listinfo/cfar-seminar
--===============0642952306==--
Technical Details
Message-ID:
<742674942D01FE498AAE0B2CB5F0FC7C0721BDEA@franklin.pc.umiacs.umd.edu>Path:
arc1-ca.pugleaf.net!g2news2.google.com!news1.google.com!news.glorb.com!npeer.de.kpn-eurorings.net!npeer-ng2.kpn.DE!uio.no!elk.ncren.net!dailyplanet.umd.edu!cs.umd.edu!mailnullShow Full Headers
Complete Article Headers
X-Google-Thread: 10f4d5,47d87597bf008845,start
X-Google-Attributes: gid10f4d5,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Path: arc1-ca.pugleaf.net!g2news2.google.com!news1.google.com!news.glorb.com!npeer.de.kpn-eurorings.net!npeer-ng2.kpn.DE!uio.no!elk.ncren.net!dailyplanet.umd.edu!cs.umd.edu!mailnull
From: janice@umiacs.umd.edu ("Janice Perrone")
Newsgroups: csd.grad
Subject: [CFAR Seminars] REMINDER: CVL Seminar - Mon. Nov. 24 at 11:00 am in
Room 2120. Speaker: Allen Yang from UC Berkeley
Date: Mon, 24 Nov 2008 10:00:40 -0500
Organization: University of Maryland College Park
Lines: 391
Sender: sendmail@dailyplanet.umd.edu
Message-ID: <742674942D01FE498AAE0B2CB5F0FC7C0721BDEA@franklin.pc.umiacs.umd.edu>
NNTP-Posting-Host: localhost
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0642952306=="
X-Trace: dailyplanet.umd.edu 1227538878 20359 127.0.0.1 (24 Nov 2008 15:01:18 GMT)
X-Complaints-To: abuse@umd.edu
NNTP-Posting-Date: Mon, 24 Nov 2008 15:01:18 +0000 (UTC)
Content-class: urn:content-classes:message
X-Received: from sausage.cs.umd.edu (server-nat-4.cs.umd.edu [128.8.127.147])
by dailyplanet.umd.edu (8.14.1/8.14.1) with ESMTP id mAOF1Fjq020352
for <csd-grad@dailyplanet.umd.edu>; Mon, 24 Nov 2008 10:01:15 -0500 (EST)
X-CSD-MailScanner-Watermark: 1228143663.69047@6ov6Bb50zUCk/SGcLUcfnQ
X-Received: from mcfeely.cs.umd.edu (mcfeely.cs.umd.edu [128.8.128.218])
by sausage.cs.umd.edu (8.13.1/8.14.1) with ESMTP id mAOF12xu006900
(version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO)
for <csd-grad@dailyplanet.umd.edu>; Mon, 24 Nov 2008 10:01:02 -0500
X-CSD-MailScanner-Watermark: 1228143662.37631@4yMhsR9qsitBwupXcmnejg
X-Received: from mcfeely.cs.umd.edu (localhost [127.0.0.1])
by mcfeely.cs.umd.edu (8.13.1/8.14.1) with ESMTP id mAOF10We003344
for <csd-grad@dailyplanet.umd.edu>; Mon, 24 Nov 2008 10:01:00 -0500
X-Received: (from mailnull@localhost)
by mcfeely.cs.umd.edu (8.13.1/8.13.1/Submit) id mAOF10KF003343
for csd-grad@news.umd.edu; Mon, 24 Nov 2008 10:01:00 -0500
X-CSD-MailScanner-Watermark: 1228143660.512@tG0hf4G/455ygFvc6UuZMQ
X-Received: from sausage.cs.umd.edu (sausage.cs.umd.edu [172.24.3.38])
by mcfeely.cs.umd.edu (8.13.1/8.14.1) with ESMTP id mAOF0xug003328;
Mon, 24 Nov 2008 10:00:59 -0500
X-CSD-MailScanner-Watermark: 1228143651.03145@TeoKNdHulT5GqDdClD42Xw
X-Received: from claven.umiacs.umd.edu (claven.umiacs.umd.edu [128.8.120.25])
by sausage.cs.umd.edu (8.13.1/8.14.1) with ESMTP id mAOF0hgH006839;
Mon, 24 Nov 2008 10:00:43 -0500
X-Received: from claven.umiacs.umd.edu (localhost [127.0.0.1])
by claven.umiacs.umd.edu (Postfix) with ESMTP id E2582191C1;
Mon, 24 Nov 2008 10:00:42 -0500 (EST)
X-Original-To: cfar-seminar@mailman.umiacs.umd.edu
X-Delivered-To: cfar-seminar@mailman.umiacs.umd.edu
X-Received: from mrouter1.umiacs.umd.edu (mrouter1.umiacs.umd.edu [128.8.120.6])
by claven.umiacs.umd.edu (Postfix) with ESMTP id 1DD39191C1
for <cfar-seminar@mailman.umiacs.umd.edu>;
Mon, 24 Nov 2008 10:00:42 -0500 (EST)
X-Received: from franklin.pc.umiacs.umd.edu (franklin.pc.umiacs.umd.edu
[128.8.118.14])
by mrouter1.umiacs.umd.edu (Postfix) with ESMTP id 19BC324C185
for <cfar-seminar@cfar.umd.edu>; Mon, 24 Nov 2008 10:00:42 -0500 (EST)
X-MimeOLE: Produced By Microsoft Exchange V6.5
X-Thread-Topic: REMINDER: CVL Seminar - Mon. Nov. 24 at 11:00 am in Room 2120.
Speaker: Allen Yang from UC Berkeley
X-Thread-Index: AclIu3ItNc06eGU2TPW/CWthGHiRrwBrVSOQAPcmkIA=
X-To: <cfar-seminar@cfar.umd.edu>
X-BeenThere: cfar-seminar@mailman.umiacs.umd.edu
X-Mailman-Version: 2.1.9
X-Precedence: list
X-List-Id: CFAR seminars <cfar-seminar.mailman.umiacs.umd.edu>
X-List-Unsubscribe: <http://mailman.umiacs.umd.edu/mailman/listinfo/cfar-seminar>,
<mailto:cfar-seminar-request@mailman.umiacs.umd.edu?subject=unsubscribe>
X-List-Post: <mailto:cfar-seminar@mailman.umiacs.umd.edu>
X-List-Help: <mailto:cfar-seminar-request@mailman.umiacs.umd.edu?subject=help>
X-List-Subscribe: <http://mailman.umiacs.umd.edu/mailman/listinfo/cfar-seminar>,
<mailto:cfar-seminar-request@mailman.umiacs.umd.edu?subject=subscribe>
X-Errors-To: cfar-seminar-bounces@claven.umiacs.umd.edu
X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.0 (sausage.cs.umd.edu [172.24.3.38]); Mon, 24 Nov 2008 10:01:02 -0500 (EST)
X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.0 (sausage.cs.umd.edu [172.24.3.38]); Mon, 24 Nov 2008 10:00:50 -0500 (EST)
X-CSD-MailScanner-Information: Please email staff@cs.umd.edu for more information
X-MailScanner-ID: mAOF0hgH006839
X-CSD-MailScanner: Found to be clean
X-CSD-MailScanner-SpamCheck: not spam, SpamAssassin (not cached, score=0.001,
required 5, autolearn=not spam, HTML_MESSAGE 0.00)
X-CSD-MailScanner-From: cfar-seminar-bounces@mailman.umiacs.umd.edu
X-Spam-Status: No