ÿþ<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <!-- saved from url=(0040)http://www.cs.umd.edu/~djacobs/pubs.html --> <HTML xmlns:st1 xmlns:o> <HEAD> <TITLE>David Jacobs: Selected Publications</TITLE> <META http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <META content="David Jacobs" name="Author"> <META content="Microsoft FrontPage 12.0" name="GENERATOR"> </HEAD> <BODY> <H2><A href="http://www.cs.umd.edu/~djacobs/">David W. Jacobs</A>: Selected Publications</H2> <H3>Newer Stuff</H3> <ul> <li> Abhishek Sharma, David Jacobs, <a href="pubs_files/PLSCVPR.pdf"> Bypassing Synthesis: PLS for Face Recognition with Pose, Low-Resolution and Sketch, </a> IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011. </li> <li> Carlos Castillo, David Jacobs, <a href="pubs_files/SlantCVPRFinal.pdf"> Face Recognition with Large Pose Variation, </a> IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011. </li> <li> Anne Jorstad, David Jacobs,&nbsp; and Alain Trouve <a href="pubs_files/LightingMetricCVPRFinal.pdf">&#8220;A Deformation and Lighting Insensitive Metric for Face Recognition Based on Dense Correspondence&#8221;,</a> IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011. </li> <li> Daozheng Chen, Mustafa Bilgic, Lise Getoor, and David Jacobs, <a href="pubs_files/EDPAV_PAMI_acceptance.pdf">&#8220;Dynamic Processing Allocation in Video,&#8221;</a> IEEE Trans. on Pattern Analysis and Machine Intelligence, (accepted for publication). </li> </ul> </SPAN> <H3>Face Recognition</H3> <ul> <li> Haibin Ling, Stefano Soatto, Narayanan Ramanathan, and David Jacobs, <a href="pubs_files/face-aging-tifs.pdf">&#8220;Face Verification Across Age Progression Using Discriminative Methods,&#8221;</a> IEEE Transactions on Information Forensics and Security, 5(1):82-91, 2010. </li> <li> Carlos D. Castillo, David Jacobs, <a href="pubs_files/TPAMI_Proofs.pdf">&#8220;Using Stereo Matching with General Epipolar Geometry for 2-D Face Recognition Across Pose&#8221;</a>, IEEE Trans. on Pattern Analysis and Machine Intelligence, 31(12):2298-2304, 2009. </li> <li> Raghuraman Gopalan, David Jacobs, &#8220;<a href="pubs_files/CVIU_lighting_final.pdf">Comparing and Combining Lighting Insensitive Approaches for Face Recognition,&#8221;</a> Computer Vision and Image Understanding, 114:(1)135-145, 2010. </li> <li><SPAN style="FONT-SIZE: 14pt"><FONT size="3">Carlos D. Castillo and David W. Jacobs,<A href="www.cs.umd.edu/~djacobs/pubs_files/FaceStereoCVPR.pdf">"Using Stereo Matching for 2-D Face Recognition Across Pose"</A>, IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2007.</FONT> </SPAN></li> </ul> <H3>Biodiversity</H3> <ul> <li> Peter N. Belhumeur, Daozheng Chen, Steven Feiner, David Jacobs, W. John Kress, Haibin Ling, Ida Lopez, Ravi Ramamoorthi, Sameer Sheorey, Sean White, and Ling Zhang, <a href = "pubs_files/eccvshort.pdf">``Searching the world&#8217;s herbaria: a system for visual identification of plant species&#8221;,</a> European Conference on Computer Vision (ECCV), pp. 116-129 2008. </li> <li><SPAN style="FONT-SIZE: 14pt"><FONT size="3">Gaurav Agarwal, Peter Belhumeur, Steven Feiner, <st1:PersonName w:st="on"> David Jacobs</st1:PersonName> , W. John Kress, Ravi Ramamoorthi,&nbsp; Norman A. Bourg, Nandan Dixit, Haibin Ling, Dhruv Mahajan, Rusty Russell, Sameer Shirdhonkar,&nbsp;Kalyan Sunkavalli,&nbsp;and Sean White <A href="http://www.cs.umd.edu/~djacobs/pubs_files/TaxonFinal.pdf">``First Steps Toward an Electronic Field Guide for Plants,''</A>, <EM>Taxon, </EM>55(3): 597-610, August, 2006.<EM>&nbsp; </EM></FONT></SPAN> </li> </ul> <H3>Memory and Learning</H3> <MENU> <LI> D. Jacobs, B. Rokers, A. Rudra, and Z. Liu Completion in Humans and Machines'', </a>NIPS, 2001.<BR> A model of human memory, with experiments on human subjects. Considers simple word memory problems like: '*a*r*l*' (think of a word with an 'a' in it, followed by an 'r', followed by an 'l', with any number of intervening letters between these letters). <a href="cues_freq_red_2.zip">data</a> <LI> D. Jacobs, D. Weinshall, Y. Gdalyahu, <a href="pubs_files/non-metric.pdf">``Class Representation and Image Retrieval with Non-Metric Distances''.</a>&nbsp;&nbsp;&nbsp; <I>IEEE Trans. on Pattern Analysis and Machine Intelligence. 22(6):583-600, 2000.&nbsp; </I>Copyright (c) 2000 by IEEE. <BR> This work considers supervised learning when images are related by distances that don't obey the triangle inequality, such as a distance that considers the centaur to be similar to the man and horse, but considers the man to be very different from the horse. <BR> <IMG height="268" src="pubs_files/centaur.jpg" width="494"> <BR> </LI> </MENU> <h3>IIllumination (and Recognition)</h3> <UL> <LI> <SPAN style="FONT-SIZE: 11pt; FONT-FAMILY: 'Times New Roman'"> <SPAN style="mso-list: Ignore"> <SPAN style="FONT: 7pt 'Times New Roman'">&nbsp;</SPAN> </SPAN> <FONT size="3"></FONT> Xue Mei, Haibin Ling, David Jacobs, <a href="pubs_files/Shadows.pdf">&#8220;Sparse Representations of Cast Shadows via l1-Regularized Least Squares&#8221;,</a> IEEE International Conference on Computer Vision (ICCV), 2009. </LI> <LI> <SPAN style="FONT-SIZE: 11pt; FONT-FAMILY: 'Times New Roman'"> <SPAN style="mso-list: Ignore"> <SPAN style="FONT: 7pt 'Times New Roman'">&nbsp;</SPAN> </SPAN> <FONT size="3">M. Osadchy, D. Jacobs, M. Lindenbaum, <A href="http://www.cs.umd.edu/~djacobs/pubs_files/CombinePamiFinal.pdf"> ``Surface Dependent Representations for Illumination Insensitive Image Comparison,''</A> <I style="mso-bidi-font-style: normal">IEEE Trans. on Pattern Analysis and Machine Intelligence</I>, <B style="mso-bidi-font-weight: normal">29</B>(1):98-111, 2007.</FONT></SPAN> &nbsp; Copyright (c) 2007 by IEEE. <LI> <SPAN style="FONT-SIZE: 11pt; FONT-FAMILY: 'Times New Roman'"> <FONT size="3">R. Basri, <st1:PersonName w:st="on">D. Jacobs</st1:PersonName>, I. Kemelmacher, <A href="http://www.cs.umd.edu/~djacobs/pubs_files/PSPami.pdf"> ``Photometric Stereo with General, Unknown Lighting,''</A> <I style="mso-bidi-font-style: normal"> Int'l J. of Computer Vision.<SPAN style="mso-spacerun: yes">&nbsp; </SPAN></I><B style="mso-bidi-font-weight: normal">72</B>(3): 239-257, 2007. <o:p></o:p></FONT></SPAN> <LI>M. Osadchy, D. Jacobs, M. Lindenbaum <A href="pubs_files/lighting_equivalence.pdf">``On the Equivalence of Common Approaches to Lighting Insensitive Recognition''</A> ICCV 2005 Copyright (c) 2005 by IEEE. <LI> Sameer Shirdhonkar and David Jacobs, <a href="pubs_files/SDP_TR.pdf"> &#8220;Non-Negative Lighting and Specular Object Recognition,&#8221;</a></abbr> IEEE International Conference on Computer Vision (ICCV), Vol. II, pp. 1323-2330, 2005. (TR Version) </LI> <LI>M. Osadchy, M. Lindenbaum, D. Jacobs, <A href="pubs_files/eccv2004Rita11.pdf">``Whitening for Photometric Comparison of Smooth Surfaces under Varying Illumination,"</A> ECCV 2004. <LI>R. Osadchy, D. Jacobs, R. Ramamoorthi, <A href="pubs_files/using_spec-10.pdf">``Using Specularities for Recognition''</A>, ICCV 2003 Copyright (c) 2003 by IEEE.&nbsp; This paper is about recognizing shiny objects like:</LI> </UL> <P><IMG height="258" src="pubs_files/pubs.h1.gif" width="347" border="0"></P> <UL> <LI> R. Basri and D. Jacobs, a href="pubs_files/Basri%20Jacobs%20-%20Lambertian%20Reflectance%20and%20Linear%20Subspaces.pdf"> ``Lambertian Reflectance and Linear Subspaces'', </a>NEC Research Institute Technical Report 2000-172R, Weizmann Institute of Science Technical Report MCS00-21. A version also appeared in ICCV '01, and in PAMI 2003. <BR> We prove that the set of all reflectance functions (the mapping from surface normals to intensities) produced by Lambertian objects under distant, isotropic lighting lies close to a 9D linear subspace. This explains previous empirical results and leads to new recognition algorithms.</LI></UL> <P><a href="pubs_files/iccv-talk-01.ppt">PowerPoint version of my ICCV talk.</a></P> <P> <MENU> <LI> R. Basri and D. Jacobs, <a href="pubs_files/Basri%20Jacobs%20-%20Photometric%20Stereo%20with%20General%20Unknown%20Lighting.pdf"> ``Photometric Stereo with General, Unknown Lighting,"</a><I> IEEE Conference on Computer Vision and Pattern Recognition 2001. </I>Copyright (c) 2001 by IEEE <BR> This is a companion to the paper above, and shows how to use these results to build 3D models using 2D images. <li> H. Chen, P. Belhumeur and D. Jacobs <a href="pubs_files/cvpr00v7.pdf">"In search of Illumination Invariants"</a><i>, IEEE Conference on Computer Vision and Pattern Recognition, pp. 254-261, 2000.&nbsp; </i> Honorable Mention, Best Paper Award.&nbsp; Copyright (c) 2000 by IEEE. <LI> D. Jacobs, P. Belhumeur and R. Basri, <a href="pubs_files/cvpr98.pdf">``Comparing Images Under Variable Illumination'', </a><I>IEEE Conference on Computer Vision and Pattern Recognition, pp. 610-617, 1998. </I>Copyright (c) 1998 by IEEE <BR> This work considers the problem of telling whether two images come from the same object with a change in lighting conditions, or from two different objects: <P><IMG height="160" src="pubs_files/arohrlexs9.jpg" width="128">&nbsp;<IMG height="160" src="pubs_files/arohrlexs1.jpg" width="128"> <BR> &nbsp; <BR> &nbsp; </P> </LI> </MENU> <h3>Recognition with Regions </h3> <ul> <li> R. Basri and D. Jacobs. <a href="pubs_files/Basri%20Jacobs%20-%20Projective%20Alignment%20with%20Regions.pdf"> ``Projective Alignment with Regions,''</a><i> IEEE Trans. on Pattern Analysis and Machine Intelligence</i>, 23(5):519--527, 2001.</li> </ul> <MENU> <LI> D. Jacobs and R. Basri, <a href="pubs_files/Jacobs%20Basri%20-%203D%20to%202D%20Pose%20Determination%20with%20Regions.pdf"> ``3-D to 2-D Recognition with Regions,''</a> <I>International Journal of Computer Vision</I>, <font FACE="Times New Roman" SIZE="3"><b>34</b></font><font FACE="Times New Roman" SIZE="3">(2/3), pages 123-145, 1999.</font> <BR> <li> R. Basri and D. Jacobs, <A href="http://www.neci.nj.nec.com/homepages/dwj/min-matching-15.ps"> ``</A><a href="pubs_files/Basri%20Jacobs%20-%20Recognition%20Using%20Region%20Correspondence.pdf">Recognition Using Region Correspondences'', </a><I>International Journal of Computer Vision,</I> <B><I>25</I></B>(2):141-162, 1996.&nbsp;&nbsp; <BR> These papers describe the use of region information to determine object pose.&nbsp; The method allows the use of 2D or 3D regions of arbitrary convex shape, in the presence of arbitrary occlusion. <P><A href="http://www.neci.nj.nec.com/tr/ieee-95.html"><IMG height="243" src="pubs_files/bic-can-f.jpg" width="320"></A>&nbsp;<IMG height="243" src="pubs_files/bic-can-b.jpg" width="320"> <BR> &nbsp; </P> <li> <P>R. Basri, A. Grove, and D. Jacobs, <a href="pubs_files/Basri%20Grove%20Jacobs%20-%20Efficient%20Determiniation%20of%20Shape.pdf"> ``Efficient Determination of Shape from Multiple Images Containing Partial Information,"</a> <I>Pattern Recognition</I>, <B><I>31</I></B>(11):1691-1703, 1998.&nbsp; </P> </li> </MENU> <h3>Other Recognition </h3> <MENU> <li> <p ALIGN="left"> Ronen Basri, Pedro Felzenszwalb, Ross Girshick, David Jacobs, Caroline Klivans, <a href="pubs_files/VisibilityPublished.pdf"> ``Visibility Constraints on Features of 3D Objects,&#8217;&#8217;</a> IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009. </li> <li> <p ALIGN="left"> Sameer Shirdhonkar and David Jacobs, <a href="pubs_files/WEMD-TR.pdf">``Approximate earth mover&#8217;s distance in linear time&#8221;</a>, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008 (TR Version). </li> <li> <p ALIGN="left">Li Yi and David Jacobs, <a href="pubs_files/SilhouetteConsistency_CVPR07.pdf"> &#8220;Efficiently Determining Silhouette Consistency,&#8221;</a> IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2007. </li> <LI> <p ALIGN="left">H. Ling, D. Jacobs <A href="http://www.cs.umd.edu/~djacobs/pubs_files/ID-pami-8.pdf"> ``Shape&nbsp;Classification&nbsp;Using the Inner-Distance,''</A> IEEE Trans on Pattern Anal. and Mach. Intell. (<B>PAMI</B>), 29(2):286-299, 2007.&nbsp; Copyright (c) 2007 by IEEE.&nbsp; </p> <LI> <p ALIGN="left">H. Ling, D. Jacobs <A href="pubs_files/deformations_iccv.pdf">``Deformation Invariant Image Matching,''</A> ICCV 2005 Copyright (c) 2005 by IEEE.<IMG alt="" src="file:///C:\dwj\write\working\deformations\figures\rd1.jpg"></p> <LI> <p ALIGN="left">R. Basri, D. Roth, D. Jacobs, <a href="pubs_files/Basri%20Roth%20Jacobs%20-%20Clustering%20Appearances%20of%203D%20Objects.pdf"> ``Clustering Appearances of 3D Objects''</a>,&nbsp;<i> IEEE Conference on Computer Vision and Pattern Recognition</i>, pp.{414--420}, 1998.</p> <LI> R. Basri, L. Costa, D. Geiger, and D. Jacobs,&nbsp; <a href="pubs_files/Basri%20etal%20-%20Determining%20the%20Similarity%20of%20Deformable%20Shapes.pdf"> ``Determining the Similarity of Deformable Shapes," </a><I>Vision Research</I>, <B><I>38</I></B>:2365-2385, 1998. <LI> T. Alter and D. Jacobs, ``Uncertainty Propagation in Model-Based Recognition", <I>International Journal of Computer Vision <B>27</B></I>(2):127-159, 1998.&nbsp; Email me for a reprint.&nbsp; An earlier version appeared as: ``Error Propagation in Full 3D-from-2D Object Recognition'', <I>IEEE Conf. on Computer Vision and Pattern Recognition. pp.892--898, 1994.</I> &nbsp;&nbsp; <LI> R. Basri and D. Jacobs, <a href="pubs_files/Basri%20Jacobs%20-%20Constancy%20and%20Similarity.pdf"> ``Constancy and Similarity,"</a> <I>Computer Vision, Graphics, and Image Processing: Image Understanding</I>, <B><I>65</I></B>(3)447-449, 1997. <LI> D. Jacobs, ``Recognizing 3-D Objects Using 2-D Images'', <I><a href="ftp://publications.ai.mit.edu/ai-publications/1000-1499/AITR-1416.ps.Z"> MIT AI Lab Technical Report 1416. </a></I>This is my PhD. thesis. It is available from MIT. A paper describing the portion of my thesis dealing with indexing 3-D objects from 2-D images also appears in: "Matching 3-D Models to 2-D Images," the <I>International Journal of Computer Vision</I>, (<I><B>21</B>)1/2</I>:123--153, January, 1997.&nbsp; Email me for a reprint. <LI> D. Jacobs, ``The Space Requirements of Indexing Under Perspective Projection'', <I>IEEE Trans. on Pattern Analysis and Machine Intelligence </I> 18(3):{330--333}, March, 1996.&nbsp; Email me for a reprint. <LI> E. Grimson, D. Huttenlocher, and D. Jacobs, ``Affine Matching with Bounded Sensor Error: A Study of Geometric Hashing and Alignment'', <I>International Journal of Computer Vision. 13(1):7--32, 1994.</I> <LI> D. Clemens and D. Jacobs, ``Space and Time Bounds on Indexing 3D Models from 2D Images'', <I>IEEE Trans. on Pattern Analysis and Machine Intelligence. 13(10):1007--1017, 1991.&nbsp; </I>Email me for a reprint.</LI></MENU> <H3>Structure from Motion</H3> <MENU> <LI> D. Jacobs, <a href="pubs_files/cviu1999-0058.pdf">``Linear Fitting with Missing Data for Structure-from-Motion'',&nbsp;</a><i> <font FACE="Times New Roman" SIZE="3"> Computer Vision and Image Understanding</font></i><font FACE="Times New Roman" SIZE="3">,</font> <font FACE="Times New Roman" SIZE="3"><b>82</b></font><font FACE="Times New Roman" SIZE="3">:57-81, 2001.</font> <LI> D. Jacobs and C. Chennubhotla, ``Segmenting Independently Moving, Noisy Points'', <I>IEEE Workshop on Motion of Non-Rigid and Articulate Objects. pp.96--103, 1994.</I> <LI> D. Jacobs and C. Chennubhotla, ``Finding Structurally Consistent Motion Correspondences'', <I>12th Int. Conf. on Pattern Recognition. pp.650--653, 1994.</I></LI> </MENU> <H3>Perceptual Organization</H3> <UL> <LI> <FONT face="Times New Roman">D. Jacobs, <A href="pubs_files/draft4-3.pdf">``What Makes Viewpoint Invariant Properties Perceptually Salient?''</A>, JOSA (A), 2003. </FONT> <LI> Z. Liu, D. Jacobs, R. Basri <a href="http://www.sciencedirect.com/science?_ob=ArticleURL&amp;_udi=B6T0W-3XR2RHW-C&amp;_user=4423&amp;_handle=W-WA-A-A-AV-MsSAYWW-UUW-AUCCEVVAZW-AUBUDDCEC-AV-U&amp;_fmt=full&amp;_coverDate=12%2F31%2F1999&amp;_rdoc=11&amp;_orig=browse&amp;_srch=%23toc%234873%231999%23999609974%23137210!&amp;_cdi=4873&amp;view=c&amp;_acct=C000000152&amp;_version=1&amp;_urlVersion=0&amp;_userid=4423&amp;md5=caba17fc987f46233eba8810ff7270a5"> ``The role of convexity in perceptual completion -- beyond good continuation''</a>. <i>Vision Research</i> 39: 4244-4257, 1999.&nbsp; </LI> </UL> <MENU> <LI> D. Jacobs, <a href="pubs_files/convex-grouping.pdf">``Robust and Efficient Detection of Convex Groups''</a>, <I>IEEE Trans. PAMI, <B>18</B></I>(1):23-37, 1996.<I> Copyright (c) 1996 by IEEE.</I> <LI> L. Williams, and D. Jacobs <a href="pubs_files/companion2.pdf">``Local Parallel Computation of Stochastic Completion Fields'' </a>,&nbsp; <I>Neural Computation</I>,<B><I> 9</I></B>(4):859-881, 1997.&nbsp;&nbsp; <LI> L. Williams and D. Jacobs <a href="pubs_files/nc2.pdf">``Stochastic Completion Fields: A Neural Model of Illusory Contour Shape and Salience'' </a>, <I>Neural Computation</I>, <B><I>9</I></B>(4):837-858, 1997.&nbsp; <LI> D. Jacobs, ``GROPER: A Grouping Based Object Recognition System for Two-Dimensional Objects'', <I>IEEE Workshop on Computer Vision. pp.164--169, 1987. Also, <a href="pubs_files/masters.pdf">MIT AI Memo 1177.</a></I> </LI> </MENU> <H3>Graphics and User Interfaces</H3> <LI> Youngmin Kim, Amitabh Varshney, David Jacobs, and Francois Guimbretiere, <A href="pubs_files/tap07.pdf">Mesh Saliency and Human Eye Fixations</A>, <i>ACM Transactions on Applied Perception</i>, (2):12:1-12:13, 2009. <LI> C. H. Lee, A. Varshney, D. Jacobs, <A href="http://www.cs.umd.edu/~djacobs/pubs_files/mesh_saliency_sig05.pdf"> ``Mesh Saliency''</A> SIGGRAPH 2005.<LI> B. Suh, H. Ling, B. Bederson, D. Jacobs <A href="pubs_files/UIST2003.pdf">``Automatic Thumbnail Cropping and its Effectiveness,''</A> UIST 2003, received best student paper award. <LI> T. Funkhouser, P. Min, M. Kazhdan, J. Chen, A. Halderman, D. Dobkin, D. Jacobs, <A href="pubs_files/tog03.pdf"> ``A Search Engine for 3D Models''</A>, <FONT face="Times New Roman"><I>ACM Transactions on Graphics</I>,</FONT> <FONT face="Times New Roman">22(1), 2003.</FONT>University of Maryland <P><A href="pubs_files/CV@Maryland.html">What's going on in computer vision at the University of Maryland?</A></P> </LI> </BODY> </HTML>