[[3D Data Processing Visualization and Transmission|http://www.cs.unc.edu/Events/Conferences/3DPVT06]]
3D reconstruction of urban environments over time. In Fall 2005 made a modest start with [[4D-Atlanta]]. See also [[Fall 05 Projects]].
We are currently focused on building a small 4D-pilot project centered around the Candler building in Downtown Atlanta. Please explore the results of this seed effort using our new [[4D Cities Viewer|http://4d-cities.cc.gatech.edu/atlanta]].\n
This project is supported by a grant from the\n[img[National Science Foundation|images/nsf.gif][http://www.nsf.gov]]\nand a gift from [img[Microsoft Research|images/msr.gif][http://research.microsoft.com]]
Hover to see names, click to link to homepages:\n!!! Graduate Students:\n[img[Kevin Quennesson|images/Students/KevinQuennesson.jpg][http://www.kevinquennesson.com] [img[Jahmeilah Richardson|images/Students/JahmeilahRichardson.jpg][http://www.cc.gatech.edu/~jarichar]] [img[Craig Cambias|images/Students/CraigCambias.jpg][http://www.cc.gatech.edu/~cambias]] [img[Josh Jones|images/Students/JoshuaJones.jpg][http://www.cc.gatech.edu/~jkj]] [img[Panchapagesan Krishnamurthy|images/Students/PanchapagesanKrishnamurthy.jpg][http://www.cc.gatech.edu/~kpanch]] [img[Lan Guo|images/Students/LanGuo.jpg][http://www.cc.gatech.edu/~languo]] [img[Arvind Kumar|images/Students/ArvindKumar.jpg][http://www.cc.gatech.edu/~arvindk]] [img[Hunter McEwan|images/Students/HunterMcEwan.jpg]] [img[Jessica Abraham|images/Students/JessicaAbraham.jpg]] [img[Sangmin Lee|images/Students/SangminLee.jpg]] \n!!!Undergraduates:\n[img[Cheol-woo Jung|images/Students//Cheol-wooJung.jpg][http://www.cc.gatech.edu/~cjung]] [img[Tom Amundsen|images/Students/TomAmundsen.jpg][http://www.cc.gatech.edu/~amundsen]] [img[Royce Cho|images/Students/RoyceCho.jpg]] [img[Bobby Mathew|images/Students/BobbyMathew.jpg][http://www.prism.gatech.edu/~gtg909x]][img[Aaron St. Clair|images/Students/AaronStClair.jpg][http://www.prism.gatech.edu/~gtg111w/]] [img[Ray Marceau|images/Students/RayMarceau.jpg][http://www.prism.gatech.edu/~gtg111w/]]
Atlanta via Microsoft's [[Virtual Earth]] API:\n\n<html><div id="map" style="position:relative;width: 500; height: 400px"></div></html><script> \n map = new VEMap('map');\n map.LoadMap(new VELatLong(33.756961,-84.3878), 17, 'h', false);\n map.HideDashboard();\n importXML( 'xml/complete_data_set.xml', 'parseXML' );\n</script>\nThere is also a [[larger, standalone map|html/VirtualEarth.html]].
[[IEEE Computer Society Conference on Computer Vision and Pattern Recorgnition|http://cvpr.cv.ri.cmu.edu/]]
Frank Dellaert\n\nemail: frank @ cc.gatech.edu (remove spaces)\n\nOffice: Technology Square Research Building 231\n\nphone: (404)385-2923\nfax: (404)894-0673\n\nMailing Address:\n\nFrank Dellaert\nCollege of Computing\nGeorgia Institute of Technology\n801 Atlantic Drive\nAtlanta, GA 30332-0280
[[Welcome]]\n[[Inferring Temporal Order of Images From 3D Structure]]\n[[People]]\n[[Publications]]\n[[Press]]\n[[Microsoft Gift]]\n[[Acknowledgements]]
In Fall 2005, we created a number of teams around small projects designed to get the [[4D Cities]] project up and running. The idea is to consider urban structures as loose collections of points, lines, and planes, without comitting to any semantic interpretation yet.\n\nBelow is more detail on each of the projects:\n*[[4D Images|html/05Fall-4D.html]]: Visualizing images in space and time.\n*[[Lines|html/05Fall-Lines.html]]: 3D line reconstruction based on a C. J. Taylor paper\n*[[Planes|html/05Fall-Planes.html]]: Oversegment images, group inside and between images, and reconstruct planelets.\n*[[4D Samples|html/05Fall-Sampler.html]]: Given only scan information about image content, can we locate the image in space ?\n*[[Sift|html/05Fall-Sift.html]]: Automatically important images into an existing database using SIFT features.\n*[[4D UI Design|html/05Fall-UI.html]]: Investigate the user interface issues regarding semi-automatic 4D reconstruction.\n*[[Points|html/05Fall-Points.html]]: Do guided matching given known camera locations.\n*[[Examining GIS|html/05Fall-GIS.html]]: Can we leverage existing GIS technology to our advantage ?
[[Google Maps|http://www.google.com/apis/maps/]] provides a very simple JavaScript API to its mapping servers. We use to use iether that or the [[Virtual Earth]] API to provide navigation capabilities in our database of temprally and spatially registered images.\n\nHere is the 4D Atlanta area via Google maps:\n\n<html><div id="gmap" style="width: 100%; height: 300px"></div></html><script> \n var gmap = new GMap(document.getElementById("gmap"));\n gmap.addControl(new GSmallMapControl());\n gmap.centerAndZoom(new GPoint(-84.388,33.756761), 1);\n gmap.setMapType(G_SATELLITE_TYPE);\n</script>\n
<<GS>> presented a [[paper|http://www.cc.gatech.edu/~phlosoft/files/schindler07cvpr.pdf]] at the 2007 [[CVPR]] conference on automatically inferring the temporal order of images from recovered 3D structure. Check out the movies below.\n<html>\n<br>\n<center>\n<object width="425" height="350"> <param name="movie" value="http://www.youtube.com/v/MS44gKoUAM0"> </param> <embed src="http://www.youtube.com/v/MS44gKoUAM0" type="application/x-shockwave-flash" width="425" height="350"> </embed> </object>\n</center>\n<br>\nAdditional (high-res quicktime) movies:\n<br><br>\n<center>\n<a href = "http://www.cc.gatech.edu/~phlosoft/mov/01_Algorithm_Visualization.mov"><img src = "http://www.cc.gatech.edu/~phlosoft/img/cvpr07_4.jpg" border=1 alt="Temporal Inference"></a>\n<a href = "http://www.cc.gatech.edu/~phlosoft/mov/03_4D_Flythrough.mov"><img src = "http://www.cc.gatech.edu/~phlosoft/img/cvpr07_fly.jpg" border=1 alt="4D Flythrough"></a>\n<a href = "http://www.cc.gatech.edu/~phlosoft/mov/02_Ordering_20_Images.mov"><img src = "http://www.cc.gatech.edu/~phlosoft/img/cvpr07_order.jpg" border=1 alt="Ordering 20 images"></a>\n</center>\n</html>
A client-side scripting language that works in most modern browsers and underlies dynamic HTML. Everything you see on this page is dynamically generated HTML via JavaScript.
\n[[Welcome]]\n[[People]]\n[[4D-Atlanta]]\n[[Press]]\n[[Publications]]\n[[Related Projects]]\n[[Acknowledgements]]\n[[Frank Dellaert|http://www.cc.gatech.edu/~dellaert]]\n\n<<rollover im1 images/Kevin/1.jpg images/Kevin/2.jpg http://4d-cities.cc.gatech.edu/atlanta width=120>>\n\n<<defaultView>>\n<<newTiddler>>\n<<newJournal "DD MMM YYYY">>\n\n<<alias FD "[[Frank Dellaert|http://www.dellaert.cc.gatech.edu]]">>\n<<alias QK "[[Kevin Quennesson|http://www.kevinquennesson.com]]">>\n<<alias GS "[[Grant Schindler|http://www.cc.gatech.edu/~phlosoft]]">>\n<<alias SB "[[Sing Bing Kang|research.microsoft.com/~sbkang]]">>\n<<alias PK "[[Panch Krishnamurthy|http://www.cc.gatech.edu/~kpanch]]">>\n\n\n<<counter>>\n
We were very lucky to be the recipient of an academic research gift from Microsoft, see the [[Microsoft Press Release|http://www.microsoft.com/presspass/press/2006/feb06/02-16TWCMapPointRFPR.mspx]]. Quoting from it:\n\nREDMOND, Wash. — Feb. 16, 2006 — Microsoft Corp. today announced the recipients of approximately $1 million in academic research funding. Through a request for proposal (RFP) process, Microsoft is encouraging academic research focused on advancing Microsoft® Virtual Earth™ technology as well as developing Trustworthy Computing curriculum projects. The 23 grant recipients represent universities from countries around the world, including in Belgium, India, Russia, South Korea and the United States. The maximum individual grant amount for each RFP is $50,000 (U.S.).\n...\nThe Virtual Earth RFP, initiated and funded by Microsoft’s Virtual Earth and Local Search business units, is designed to encourage university research in areas relevant to digital geography, including spatio-temporal databases, routing, computer vision, ontologies, map user interfaces and visualization.\n...\nThe eight winners of the Virtual Earth RFP will conduct basic research in digital geographics that is expected to advance the state of the art.
[[Microsoft Research|http://research.microsoft.com]] is the research lab of [[Microsoft|www.microsoft.com]]. Take a look at their [[Interactive Visual Media Group|http://research.microsoft.com/vision/InteractiveVisualMediaGroup]].
[img[NSF|http://www.nsf.gov/images/head.gif]]\n[[National Science Foundation|http://www.nsf.gov]]
The 4D Cities project is lead by [[Prof. Frank Dellaert|http://www.cc.gatech.edu/~dellaert]]. Students working on the project in 2007 are listed below (hover to see names, click to link to homepages):\n\n[img[Dan Hou|images/Students/DanHou.jpg]] [img[Mingxuan Sun|images/Students/MingXuanSun.jpg]] [img[Grant Schindler|images/Students/GrantSchindler.jpg][http://www.cc.gatech.edu/~phlosoft]] [img[Alistair Jones|images/Students/AlistairJones.jpg][http://www.facebook.com/profile.php?id=12802267]] \n\n[[Alumni]]
* [[New Scientist June 2007|http://technology.newscientist.com/channel/tech/mg19426106.200-see-a-city-change-in-four-dimensions.html]]\n* [[ZDNet Emerging Tech July 2007|http://blogs.zdnet.com/emergingtech/?p=627]]
* [[Detecting and Matching Repeated Patterns for Automatic Geo-tagging in Urban Environments|http://www.cc.gatech.edu/~phlosoft/files/schindler08cvpr.pdf]], <<GS>>, P. Krishnamurthy, R. Lublinerman, Y. Liu, and <<FD>>, IEEE Comp. Soc. Conf. on Computer Vision and Pattern Recognition ([[CVPR]]), 2008. \n* [[Inferring Temporal Order of Images From 3D Structure|http://www.cc.gatech.edu/~phlosoft/files/schindler07cvpr.pdf]], <<GS>>, <<FD>>, and S.B. Kang, IEEE Computer Society Conference on Computer Vision and Pattern Recognition ([[CVPR]]), 2007. \n* <<FD>>'s [[invited talk on 4D-cities|talks/2006-06-14-3DPVT.pdf]] in pdf (8.3MB).\n* <<pub conference key 2006 "Line-Based Structure From Motion for Urban Environments" "GS,PK,FD" "3DPVT" pubs/Schindler06-3dpvt.pdf>>\n* <<pub conference key 2006 "Rao-Blackwellized Importance Sampling of Camera Parameters from Simple User Input with Visibility Preprocessing in Line Space" "KQ,FD" "3DPVT" pubs/Quennesson06-3dpvt.pdf>>\n\nSee also the following relevant publications by our group:\n<html>\n<ul>\n\n <li><a href="http://www.cc.gatech.edu/%7Edellaert/ftp/Schindler04cvpr.pdf">Atlanta\nWorld: An Expectation-Maximization Framework for Simultaneous Low-level\nEdge Grouping and Camera Calibration in Complex Man-Made Environments</a>,\nCVPR 04</li>\n\n <li><a href="http://www.cc.gatech.edu/%7Edellaert/ftp/Kaess04eccv.pdf">Multiview\nReconstruction of Piecewise Smooth Subdivision Curves with a Variable\nNumber of Control Points</a>, ECCV 04</li>\n\n <li><a href="http://www.cc.gatech.edu/cpl/projects/reconstructions/ICCV03/abstract.html">Spectral\nPartitioning for Structure From Motion,</a> ICCV 03</li>\n\n <li><a href="http://www.cc.gatech.edu/%7Edellaert/pubs/ml.pdf">EM,\nMCMC, and Chain Flipping for Structure from Motion with \n\nUnknown Correspondence</a> , Machine Learning Journal, 2003</li>\n\n <li><a href="http://www.ri.cmu.edu/pubs/pub_3851.html">Monte\nCarlo EM for Data-Association and its Applications in Computer Vision</a>,\nPh.D. Thesis, 2001</li>\n\n <li><a href="http://www.ri.cmu.edu/pubs/pub_3429.html">Feature \nCorrespondence: A Markov Chain Monte Carlo Approach</a>, NIPS 00</li>\n\n <li><a href="http://www.ri.cmu.edu/pubs/pub_3292.html">Structure\nfrom Motion without Correspondence</a>, CVPR 00</li>\n\n </ul>\n</html>
* [[The city of Bologna in 4D|http://www.storiaeinformatica.it/nume/english/ntitolo_eng.html]], which takes a CAD/VR approach. They also review [[many related endaveours|http://www.storiaeinformatica.it/nume/english/narch_eng.html]].\n* The [[MIT City Scanning Project|http://city.csail.mit.edu/city.html]]: Fully Automated Model Acquisition in Urban Areas
~Spatio-Temporal Reconstruction from Images
4D Cities
#titleLine { color:#ffeecc}\n\n#displayArea {margin: 1em 12em 0em 14em}\n.toolbar{padding: 0;}\n\n/* sidebar */\n#sidebar {width: 12em;}\n#sidebarOptions {background-color: #ffffff}\n#sidebarTabs {background-color: #ffffff}\n#sidebarTabs .tabSelected {top: 0px}\n#sidebarTabs .tabContents {width: 11em; background-color: #ffffff}\n#sidebarTabs .tabContents .tiddlyLink {color: #996633}\n#sidebarTabs .tabContents .button {color: #996633}\n\n/* shift tiddler toolbar onto same line as tiddler title */\n.toolbar { float:right; display:inline; padding-bottom:0; }\n\n#tiddlerWelcome img {display: block; margin-left: auto; margin-right: auto}\n#tiddler4D-Atlanta div {display: block; margin-left: auto; margin-right: auto}\n.title {background-color: #ffeecc}\n.viewer h3 {background-color: #ffffdd}
[[Virtual Earth|http://www.viavirtualearth.com]] is Microsoft's answer to [[Google Maps]]. It's API has a cool feature in that allows you to display hybrid maps. See [[4D-Atlanta]] and the [[larger, standalone map|html/VirtualEarth.html]], both Virtual Earth maps of Atlanta annotated with some of the images we work with.
''Unlocking the Urban Photographic Record through 4D Scene Understanding''\n\n[img[result_3|images/Atlanta/4D-strip.jpg][html/05Fall-4D.html]]\nThere is a growing need for novel ways to access the exponentially growing archives of historical imagery. It is imperative to go beyond cataloging, indexing, and keyword driven databases, to a paradigm where the computer at least partially understands the content of images. Pushing the state of the art in scene understanding and 3D modeling will enable radical new ways to view and experience historical and/or temporally varying imagery. The research described here aims at building ''time-varying 3D models'' that can serve to pull together large collections of images pertaining to the appearance, evolution, and events surrounding one place or artifact over time, as exemplified by the [[4D Cities]] project: the completely automatic construction of a 4D database showing the evolution over time of a single city.\n\nThis project is partially funded by [[NSF]] and a gift by [[Microsoft Research]].