Lecture Notes in Computer Science 8689Commenced Publication in 1973Founding and Former Series Editors:Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen

Editorial Board

David HutchisonLancaster University, UK

Takeo KanadeCarnegie Mellon University, Pittsburgh, PA, USA

Josef KittlerUniversity of Surrey, Guildford, UK

Jon M. KleinbergCornell University, Ithaca, NY, USA

Alfred KobsaUniversity of California, Irvine, CA, USA

Friedemann MatternETH Zurich, Switzerland

John C. MitchellStanford University, CA, USA

Moni NaorWeizmann Institute of Science, Rehovot, Israel

Oscar NierstraszUniversity of Bern, Switzerland

C. Pandu RanganIndian Institute of Technology, Madras, India

Bernhard SteffenTU Dortmund University, Germany

Demetri TerzopoulosUniversity of California, Los Angeles, CA, USA

Doug TygarUniversity of California, Berkeley, CA, USA

Gerhard WeikumMax Planck Institute for Informatics, Saarbruecken, Germany

David Fleet Tomas Pajdla Bernt SchieleTinne Tuytelaars (Eds.)

Computer Vision –ECCV 201413th European ConferenceZurich, Switzerland, September 6-12, 2014Proceedings, Part I

13

Volume Editors

David FleetUniversity of Toronto, Department of Computer Science6 King’s College Road, Toronto, ON M5H 3S5, CanadaE-mail: fleet@cs.toronto.edu

Tomas PajdlaCzech Technical University in Prague, Department of CyberneticsTechnicka 2, 166 27 Prague 6, Czech RepublicE-mail: pajdla@cmp.felk.cvut.cz

Bernt SchieleMax-Planck-Institut für InformatikCampus E1 4, 66123 Saarbrücken, GermanyE-mail: schiele@mpi-inf.mpg.de

Tinne TuytelaarsKU Leuven, ESAT - PSI, iMindsKasteelpark Arenberg 10, Bus 2441, 3001 Leuven, BelgiumE-mail: tinne.tuytelaars@esat.kuleuven.be

Videos to this book can be accessed athttp://www.springerimages.com/videos/978-3-319-10589-5

ISSN 0302-9743 e-ISSN 1611-3349ISBN 978-3-319-10589-5 e-ISBN 978-3-319-10590-1DOI 10.1007/978-3-319-10590-1Springer Cham Heidelberg New York Dordrecht London

Library of Congress Control Number: 2014946360

LNCS Sublibrary: SL 6 – Image Processing, Computer Vision, Pattern Recognition,and Graphics

© Springer International Publishing Switzerland 2014This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part ofthe material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,broadcasting, reproduction on microfilms or in any other physical way, and transmission or informationstorage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodologynow known or hereafter developed. Exempted from this legal reservation are brief excerpts in connectionwith reviews or scholarly analysis or material supplied specifically for the purpose of being entered andexecuted on a computer system, for exclusive use by the purchaser of the work. Duplication of this publicationor parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location,in ist current version, and permission for use must always be obtained from Springer. Permissions for usemay be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecutionunder the respective Copyright Law.The use of general descriptive names, registered names, trademarks, service marks, etc. in this publicationdoes not imply, even in the absence of a specific statement, that such names are exempt from the relevantprotective laws and regulations and therefore free for general use.While the advice and information in this book are believed to be true and accurate at the date of publication,neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors oromissions that may be made. The publisher makes no warranty, express or implied, with respect to thematerial contained herein.

Typesetting: Camera-ready by author, data conversion by Scientific Publishing Services, Chennai, India

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Foreword

The European Conference on Computer Vision is one of the top conferences incomputer vision. It was first held in 1990 in Antibes (France) with subsequentconferences in Santa Margherita Ligure (Italy) in 1992, Stockholm (Sweden) in1994, Cambridge (UK) in 1996, Freiburg (Germany) in 1998, Dublin (Ireland) in2000, Copenhagen (Denmark) in 2002, Prague (Czech Republic) in 2004, Graz(Austria) in 2006, Marseille (France) in 2008, Heraklion (Greece) in 2010, andFlorence (Italy) in 2012. Many people have worked hard to turn the 2014 editioninto as great a success. We hope you will find this a mission accomplished.

The chairs decided to adhere to the classic single-track scheme. In terms ofthe time ordering, we decided to largely follow the Florence example (typicallystarting with poster sessions, followed by oral sessions), which offers a lot offlexibility to network and is more forgiving for the not-so-early-birds and hard-core gourmets.

A large conference like ECCV requires the help of many. They made surethere was a full program including the main conference, tutorials, workshops,exhibits, demos, proceedings, video streaming/archive, andWeb descriptions.Wewant to cordially thank all those volunteers! Please have a look at the conferencewebsite to see their names (http://eccv2014.org/people/). We also thank ourgenerous sponsors. Their support was vital for keeping prices low and enrichingthe program. And it is good to see such a level of industrial interest in what ourcommunity is doing!

We hope you will enjoy the proceedings ECCV 2014.

Also, willkommen in Zurich!

September 2014 Marc PollefeysLuc Van GoolGeneral Chairs

Preface

Welcome to the proceedings of the 2014 European Conference on Computer Vi-sion (ECCV 2014) that was in Zurich, Switzerland. We are delighted to presentthis volume reflecting a strong and exciting program, the result of an extensivereview process. In total, we received 1,444 paper submissions. Of these, 85 vio-lated the ECCV submission guidelines and were rejected without review. Of theremainder, 363 were accepted (26,7%): 325 as posters (23,9%) and 38 as oralpresentations (2,8%). This selection process was a combined effort of four pro-gram co-chairs (PCs), 53 area chairs (ACs), 803 Program Committee membersand 247 additional reviewers.

As PCs we were primarily responsible for the design and execution of thereview process. Beyond administrative rejections, we were not directly involvedin acceptance decisions. Because the general co-chairs were permitted to submitpapers, they played no role in the review process and were treated as any otherauthor.

Acceptance decisions were made by the AC Committee. There were 53 ACs intotal, selected by the PCs to provide sufficient technical expertise, geographicaldiversity (21 from Europe, 7 from Asia, and 25 from North America) and a mixof AC experience (7 had no previous AC experience, 18 had served as AC ofa major international vision conference once since 2010, 8 had served twice, 13had served three times, and 7 had served 4 times).

ACs were aided by 803 Program Committee members to whom papers wereassigned for reviewing. There were 247 additional reviewers, each supervised bya Program Committee member. The Program Committee was based on sug-gestions from ACs, and committees from previous conferences. Google Scholarprofiles were collected for all candidate Program Committee members and vet-ted by PCs. Having a large pool of Program Committee members for reviewingallowed us to match expertise while bounding reviewer loads. No more than ninepapers were assigned to any one Program Committee member, with a maximumof six to graduate students.

The ECCV 2014 review process was double blind. Authors did not know thereviewers’ identities, nor the ACs handling their paper(s). We did our utmostto ensure that ACs and reviewers did not know authors’ identities, even thoughanonymity becomes difficult to maintain as more and more submissions appearconcurrently on arXiv.org.

Particular attention was paid to minimizing potential conflicts of interest.Conflicts of interest between ACs, Program Committee members, and paperswere based on authorship of ECCV 2014 submissions, on their home institu-tions, and on previous collaborations. To find institutional conflicts, all authors,

VIII Preface

Program Committee members, and ACs were asked to list the Internet domainsof their current institutions. To find collaborators, the DBLP (www.dblp.org)database was used to find any co-authored papers in the period 2010–2014.

We initially assigned approximately 100 papers to each AC, based on affinityscores from the Toronto Paper Matching System and authors’ AC suggestions.ACs then bid on these, indicating their level of expertise. Based on these bids,and conflicts of interest, approximately 27 papers were assigned to each AC,for which they would act as the primary AC. The primary AC then suggestedseven reviewers from the pool of Program Committee members (in rank order)for each paper, from which three were chosen per paper, taking load balancingand conflicts of interest into account.

Many papers were also assigned a secondary AC, either directly by the PCs,or as a consequence of the primary AC requesting the aid of an AC with comple-mentary expertise. Secondary ACs could be assigned at any stage in the process,but in most cases this occurred about two weeks before the final AC meeting.Hence, in addition to their initial load of approximately 27 papers, each AC wasasked to handle three to five more papers as a secondary AC; they were expectedto read and write a short assessment of such papers. In addition, two of the 53ACs were not directly assigned papers. Rather, they were available through-out the process to aid other ACs at any stage (e.g., with decisions, evaluatingtechnical issues, additional reviews, etc.).

The initial reviewing period was three weeks long, after which reviewersprovided reviews with preliminary recommendations. Three weeks is somewhatshorter than normal, but this did not seem to cause any unusual problems. Withthe generous help of several last-minute reviewers, each paper received threereviews.

Authors were then given the opportunity to rebut the reviews, primarily toidentify any factual errors. Following this, reviewers and ACs discussed papersat length, after which reviewers finalized their reviews and gave a final recom-mendation to the ACs. Many ACs requested help from secondary ACs at thistime.

Papers, for which rejection was clear and certain, based on the reviews andthe AC’s assessment, were identified by their primary ACs and vetted by ashadow AC prior to rejection. (These shadow ACs were assigned by the PCs.)All papers with any chance of acceptance were further discussed at the ACmeeting. Those deemed “strong” by primary ACs (about 140 in total) were alsoassigned a secondary AC.

The AC meeting, with all but two of the primary ACs present, took place inZurich. ACs were divided into 17 triplets for each morning, and a different set oftriplets for each afternoon. Given the content of the three (or more) reviews alongwith reviewer recommendations, rebuttals, online discussions among reviewersand primary ACs, written input from and discussions with secondary ACs, the

Preface IX

AC triplets then worked together to resolve questions, calibrate assessments, andmake acceptance decisions.

To select oral presentations, all strong papers, along with any others putforward by triplets (about 155 in total), were then discussed in four panels, eachcomprising four or five triplets. Each panel ranked these oral candidates, usingfour categories. Papers in the two top categories provided the final set of 38 oralpresentations.

We want to thank everyone involved in making the ECCV 2014 Programpossible. First and foremost, the success of ECCV 2014 depended on the qualityof papers submitted by authors, and on the very hard work of the reviewers,the Program Committee members and the ACs. We are particularly grateful toKyros Kutulakos for his enormous software support before and during the ACmeeting, to Laurent Charlin for the use of the Toronto Paper Matching System,and Chaohui Wang for help optimizing the assignment of papers to ACs. Wealso owe a debt of gratitude for the great support of Zurich local organizers,especially Susanne Keller and her team.

September 2014 David FleetTomas PajdlaBernt Schiele

Tinne Tuytelaars

Organization

General Chairs

Luc Van Gool ETH Zurich, SwitzerlandMarc Pollefeys ETH Zurich, Switzerland

Program Chairs

Tinne Tuytelaars KU Leuven, BelgiumBernt Schiele MPI Informatics, Saarbrucken, GermanyTomas Pajdla CTU Prague, Czech RepublicDavid Fleet University of Toronto, Canada

Local Arrangements Chairs

Konrad Schindler ETH Zurich, SwitzerlandVittorio Ferrari University of Edinburgh, UK

Workshop Chairs

Lourdes Agapito University College London, UKCarsten Rother TU Dresden, GermanyMichael Bronstein University of Lugano, Switzerland

Tutorial Chairs

Bastian Leibe RWTH Aachen, GermanyPaolo Favaro University of Bern, SwitzerlandChristoph Lampert IST Austria

Poster Chair

Helmut Grabner ETH Zurich, Switzerland

Publication Chairs

Mario Fritz MPI Informatics, Saarbrucken, GermanyMichael Stark MPI Informatics, Saarbrucken, Germany

XII Organization

Demo Chairs

Davide Scaramuzza University of Zurich, SwitzerlandJan-Michael Frahm University of North Carolina at Chapel Hill,

USA

Exhibition Chair

Tamar Tolcachier University of Zurich, Switzerland

Industrial Liaison Chairs

Alexander Sorkine-Hornung Disney Research Zurich, SwitzerlandFatih Porikli ANU, Australia

Student Grant Chair

Seon Joo Kim Yonsei University, Korea

Air Shelters Accommodation Chair

Maros Blaha ETH Zurich, Switzerland

Website Chairs

Lorenz Meier ETH Zurich, SwitzerlandBastien Jacquet ETH Zurich, Switzerland

Internet Chair

Thorsten Steenbock ETH Zurich, Switzerland

Student Volunteer Chairs

Andrea Cohen ETH Zurich, SwitzerlandRalf Dragon ETH Zurich, SwitzerlandLaura Leal-Taixe ETH Zurich, Switzerland

Finance Chair

Amael Delaunoy ETH Zurich, Switzerland

Conference Coordinator

Susanne H. Keller ETH Zurich, Switzerland

Organization XIII

Area Chairs

Lourdes Agapito University College London, UKSameer Agarwal Google Research, USAShai Avidan Tel Aviv University, IsraelAlex Berg UNC Chapel Hill, USAYuri Boykov University of Western Ontario, CanadaThomas Brox University of Freiburg, GermanyJason Corso SUNY at Buffalo, USATrevor Darrell UC Berkeley, USAFernando de la Torre Carnegie Mellon University, USAFrank Dellaert Georgia Tech, USAAlexei Efros UC Berkeley, USAVittorio Ferrari University of Edinburgh, UKAndrew Fitzgibbon Microsoft Research, Cambridge, UKJanMichael Frahm UNC Chapel Hill, USABill Freeman Massachusetts Institute of Technology, USAPeter Gehler Max Planck Institute for Intelligent Systems,

GermanyKristen Graumann University of Texas at Austin, USAWolfgang Heidrich University of British Columbia, CanadaHerve Jegou Inria Rennes, FranceFredrik Kahl Lund University, SwedenKyros Kutulakos University of Toronto, CanadaChristoph Lampert IST AustriaIvan Laptev Inria Paris, FranceKyuong Mu Lee Seoul National University, South KoreaBastian Leibe RWTH Aachen, GermanyVincent Lepetit TU Graz, AustriaHongdong Li Australian National UniversityDavid Lowe University of British Columbia, CanadaGreg Mori Simon Fraser University, CanadaSrinivas Narasimhan Carnegie Mellon University, PA, USANassir Navab TU Munich, GermanyKo Nishino Drexel University, USAMaja Pantic Imperial College London, UKPatrick Perez Technicolor Research, Rennes, FrancePietro Perona California Institute of Technology, USAIan Reid University of Adelaide, AustraliaStefan Roth TU Darmstadt, GermanyCarsten Rother TU Dresden, GermanySudeep Sarkar University of South Florida, USASilvio Savarese Stanford University, USAChristoph Schnoerr Heidelberg University, GermanyJamie Shotton Microsoft Research, Cambridge, UK

XIV Organization

Kaleem Siddiqi McGill, CanadaLeonid Sigal Disney Research, Pittsburgh, PA, USANoah Snavely Cornell, USARaquel Urtasun University of Toronto, CanadaAndrea Vedaldi University of Oxford, UKJakob Verbeek Inria Rhone-Alpes, FranceXiaogang Wang Chinese University of Hong Kong, SAR ChinaMing-Hsuan Yang UC Merced, CA, USALihi Zelnik-Manor Technion, IsraelSong-Chun Zhu UCLA, USATodd Zickler Harvard, USA

Program Committee

Gaurav AggarwalAmit AgrawalHaizhou AiIjaz AkhterKarteek AlahariAlexandre AlahiAndrea AlbarelliSaad AliJose M. AlvarezJuan Andrade-CettoBjoern AndresMykhaylo AndrilukaElli AngelopoulouRoland AngstRelja ArandjelovicOgnjen ArandjelovicHelder AraujoPablo ArbelezVasileios ArgyriouAntonis ArgyrosKalle AstroemVassilis AthitsosYannis AvrithisYusuf AytarXiang BaiLuca BallanYingze BaoRichard BaraniukAdrian BarbuKobus BarnardConnelly Barnes

Joao BarretoJonathan BarronAdrien BartoliArslan BasharatDhruv BatraLuis BaumelaMaximilian BaustJean-Charles BazinLoris BazzaniChris BeallVasileios BelagiannisCsaba BeleznaiMoshe Ben-ezraOhad Ben-ShaharIsmail Ben AyedRodrigo BenensonRyad BenosmanTamara BergMargrit BetkeRoss BeveridgeBir BhanuHorst BischofArijit BiswasAndrew BlakeAaron BobickPiotr BojanowskiAli BorjiTerrance BoultLubomir BourdevPatrick BouthemyEdmond Boyer

Kristin BransonSteven BransonFrancois BremondMichael BronsteinGabriel BrostowMichael BrownMatthew BrownMarcus BrubakerAndres BruhnJoan BrunaAurelie BugeauDarius BurschkaRicardo CabralJian-Feng CaiNeill D.F. CampbellYong CaoBarbara CaputoJoao CarreiraJan CechJinxiang ChaiAyan ChakrabartiTat-Jen ChamAntoni ChanManmohan ChandrakerVijay ChandrasekharHong ChangMing-Ching ChangRama ChellappaChao-Yeh ChenDavid ChenHwann-Tzong Chen

Organization XV

Tsuhan ChenXilin ChenChao ChenLongbin ChenMinhua ChenAnoop CherianLiang-Tien ChiaTat-Jun ChinSunghyun ChoMinsu ChoNam Ik ChoWongun ChoiMario ChristoudiasWen-Sheng ChuYung-Yu ChuangOndrej ChumJames ClarkBrian ClippIsaac CohenJohn CollomosseBob CollinsTim CootesDavid CrandallAntonio CriminisiNaresh CuntoorQieyun DaiJifeng DaiKristin DanaKostas DaniilidisLarry DavisAndrew DavisonGoksel DedeogluKoichiro DeguchiAlberto Del BimboAlessio Del BueHerve DelingetteAndrew DelongStefanie DemirciDavid DemirdjianJia DengJoachim DenzlerKonstantinos DerpanisThomas DeselaersFrederic DevernayMichel Dhome

Anthony DickAjay DivakaranSantosh Kumar DivvalaMinh DoCarl DoerschPiotr DollarBin DongWeisheng DongMichael DonoserGianfranco DorettoMatthijs DouzeBruce DraperMark DrewBertram DrostLixin DuanJean-Luc DugelayEnrique DunnPinar DuyguluJan-Olof EklundhJames H. ElderIan EndresOlof EnqvistMarkus EnzweilerAykut ErdemAnders ErikssonAli EslamiIrfan EssaFrancisco EstradaBin FanQuanfu FanJialue FanSean FanelloAli FarhadiGiovanni FarinellaRyan FarrellAlireza FathiPaolo FavaroMichael FelsbergPedro FelzenszwalbRob FergusBasura FernandoFrank FerrieSanja FidlerBoris FlachFrancois Fleuret

David FofiWolfgang FoerstnerDavid ForsythKaterina FragkiadakiJean-Sebastien FrancoFriedrich FraundorferMario FritzYun FuPascal FuaHironobu FujiyoshiYasutaka FurukawaRyo FurukawaAndrea FusielloFabio GalassoJuergen GallAndrew GallagherDavid GallupArvind GaneshDashan GaoShenghua GaoJames GeeAndreas GeigerYakup GencBogdan GeorgescuGuido GerigDavid GeronimoTheo GeversBernard GhanemAndrew GilbertRoss GirshickMartin GodecGuy GodinRoland GoeckeMichael GoeseleAlvina GohBastian GoldlueckeBoqing GongYunchao GongRaghuraman GopalanAlbert GordoLena GorelickPaulo GotardoStephen GouldVenu Madhav GovinduHelmut Grabner

XVI Organization

Roger GrosseMatthias GrundmannChunhui GuXianfeng GuJinwei GuSergio GuadarramaMatthieu GuillauminJean-Yves GuillemautHatice GunesRuiqi GuoGuodong GuoAbhinav GuptaAbner Guzman RiveraGregory HagerGhassan HamarnehBohyung HanTony HanJari HannukselaTatsuya HaradaMehrtash HarandiBharath HariharanStefan HarmelingTal HassnerDaniel HauaggeSøren HaubergMichal HavlenaJames HaysKaiming HeXuming HeMartial HebertFelix HeideJared HeinlyHagit Hel-OrLionel HengPhilipp HennigCarlos HernandezAaron HertzmannAdrian HiltonDavid HoggDerek HoiemByung-Woo HongAnthony HoogsJoachim HorneggerTimothy HospedalesWenze Hu

Zhe HuGang HuaXian-Sheng HuaDong HuangGary HuangHeng HuangSung Ju HwangWonjun HwangIvo IhrkeNazli Ikizler-CinbisSlobodan IlicHorace IpMichal IraniHiroshi IshikawaLaurent IttiNathan JacobsMax JaderbergOmar JavedC.V. JawaharBruno JedynakHueihan JhuangQiang JiHui JiKui JiaYangqing JiaJiaya JiaHao JiangZhuolin JiangSam JohnsonNeel JoshiArmand JoulinFrederic JurieIoannis KakadiarisZdenek KalalAmit KaleJoni-Kristian

KamarainenGeorge KamberovKenichi KanataniSing Bing KangVadim KantorovJorg Hendrik KappesLeonid KarlinskyZoltan KatoHiroshi Kawasaki

Verena KaynigCem KeskinMargret KeuperDaniel KeysersSameh KhamisFahad KhanSaad KhanAditya KhoslaMartin KiefelGunhee KimJaechul KimSeon Joo KimTae-Kyun KimByungsoo KimBenjamin KimiaKris KitaniHedvig KjellstromLaurent KneipReinhard KochKevin KoeserUllrich KoetheEffrosyni KokiopoulouIasonas KokkinosKalin KolevVladimir KolmogorovVladlen KoltunNikos KomodakisPiotr KoniuszPeter KontschiederEnder KonukogluSanjeev KoppalHema KoppulaAndreas KoschanJana KoseckaAdriana KovashkaAdarsh KowdleJosip KrapacDilip KrishnanZuzana KukelovaBrian KulisNeeraj KumarM. Pawan KumarCheng-Hao KuoIn So KweonJunghyun Kwon

Organization XVII

Junseok KwonSimon Lacoste-JulienShang-Hong LaiJean-Francois LalondeTian LanMichael LangerDoug LanmanDiane LarlusLongin Jan LateckiSvetlana LazebnikLaura Leal-TaixeErik Learned-MillerHonglak LeeYong Jae LeeIdo LeichterVictor LempitskyFrank LenzenMarius LeordeanuThomas LeungMaxime LhuillierChunming LiFei-Fei LiFuxin LiRui LiLi-Jia LiChia-Kai LiangShengcai LiaoJoerg LiebeltJongwoo LimJoseph LimRuei-Sung LinYen-Yu LinZhouchen LinLiang LinHaibin LingJames LittleBaiyang LiuCe LiuFeng LiuGuangcan LiuJingen LiuWei LiuZicheng LiuZongyi LiuTyng-Luh Liu

Xiaoming LiuXiaobai LiuMing-Yu LiuMarcus LiwickiStephen LombardiRoberto Lopez-SastreManolis LourakisBrian LovellChen Change LoyJiangbo LuJiwen LuSimon LuceyJiebo LuoPing LuoMarcus MagnorVijay MahadevanJulien MairalMichael MaireSubhransu MajiAtsuto MakiYasushi MakiharaRoberto ManduchiLuca MarchesottiAleix MartinezBogdan MateiDiana MateusStefan MatheYasuyuki MatsushitaIain MatthewsKevin MatzenBruce MaxwellStephen MaybankWalterio Mayol-CuevasDavid McAllesterGerard MedioniChristopher MeiPaulo MendoncaThomas MensinkDomingo MeryAjmal MianBranislav MicusikOndrej MiksikAnton MilanMajid MirmehdiAnurag Mittal

Hossein MobahiPranab MohantyPascal MonasseVlad MorariuPhilippos MordohaiFrancesc Moreno-NoguerLuce MorinNigel MorrisBryan MorseEric MortensenYasuhiro MukaigawaLopamudra MukherjeeVittorio MurinoDavid MurraySobhan Naderi PariziHajime NagaharaLaurent NajmanKarthik NandakumarFabian NaterJan NeumannLukas NeumannRam NevatiaRichard NewcombeMinh Hoai NguyenBingbing NiFeiping NieJuan Carlos NieblesMarc NiethammerClaudia NieuwenhuisMark NixonMohammad NorouziSebastian NowozinMatthew O’ToolePeter OchsJean-Marc OdobezFrancesca OdoneEyal OfekSangmin OhTakahiro OkabeTakayuki OkataniAude OlivaCarl OlssonBjorn OmmerMagnus OskarssonWanli Ouyang

XVIII Organization

Geoffrey OxholmMustafa OzuysalNicolas PadoyCaroline PantofaruNicolas PapadakisGeorge PapandreouNikolaos

PapanikolopoulosNikos ParagiosDevi ParikhDennis ParkVishal PatelIoannis PatrasVladimir PavlovicKim PedersenMarco PedersoliShmuel PelegMarcello PelilloTingying PengA.G. Amitha PereraAlessandro PerinaFederico PerniciFlorent PerronninVladimir PetrovicTomas PfisterJonathon PhillipsJustus PiaterMassimo PiccardiHamed PirsiavashLeonid PishchulinRobert PlessThomas PockJean PonceGerard Pons-MollRonald PoppeAndrea PratiVictor PrisacariuKari PulliYu QiaoLei QinNovi QuadriantoRahul RaguramVarun RamakrishnaSrikumar RamalingamNarayanan Ramanathan

KonstantinosRapantzikos

Michalis RaptisNalini RathaAvinash RavichandranMichael RealeDikpal ReddyJames RehgJan ReininghausXiaofeng RenJerome RevaudMorteza RezanejadHayko RiemenschneiderTammy Riklin RavivAntonio Robles-KellyErik RodnerEmanuele RodolaMikel RodriguezMarcus RohrbachJavier RomeroCharles RosenbergBodo RosenhahnArun RossSamuel Rota BulPeter RothVolker RothAnastasios RoussosSebastien RoyMichael RubinsteinOlga RussakovskyBryan RussellMichael S. RyooMohammad Amin

SadeghiKate SaenkoAlbert Ali SalahImran SaleemiMathieu SalzmannConrad SandersonAswin

SankaranarayananBenjamin SappRadim SaraScott SatkinImari Sato

Yoichi SatoBogdan SavchynskyyHanno ScharrDaniel ScharsteinYoav Y. SchechnerWalter ScheirerKevin ScheltenFrank SchmidtUwe SchmidtJulia SchnabelAlexander SchwingNicu SebeShishir ShahMubarak ShahShiguang ShanQi ShanLing ShaoAbhishek SharmaViktoriia SharmanskaEli ShechtmanYaser SheikhAlexander ShekhovtsovChunhua ShenLi ShenYonggang ShiQinfeng ShiIlan ShimshoniTakaaki ShiratoriAbhinav ShrivastavaBehjat SiddiquieNathan SilbermanKaren SimonyanRicha SinghVikas SinghSudipta SinhaJosef SivicDirk SmeetsArnold SmeuldersWilliam SmithCees SnoekEric SommerladeAlexander

Sorkine-HornungAlvaro SotoRichard Souvenir

Organization XIX

Anuj SrivastavaIoannis StamosMichael StarkChris StaufferBjorn StengerCharles StewartRainer StiefelhagenJuergen SturmYusuke SuganoJosephine SullivanDeqing SunMin SunHari SundarGanesh SundaramoorthiKalyan SunkavalliSabine SusstrunkDavid SuterTomas SvobodaRahul SwaminathanTanveer

Syeda-MahmoodRick SzeliskiRaphael SznitmanYuichi TaguchiYu-Wing TaiJun TakamatsuHugues TalbotPing TanRobby TanKevin TangHuixuan TangDanhang TangMarshall TappenJean-Philippe TarelDanny TarlowGabriel TaubinCamillo TaylorDemetri TerzopoulosChristian TheobaltYuandong TianJoseph TigheRadu TimofteMassimo TistarelliGeorge TodericiSinisa Todorovic

Giorgos ToliasFederico TombariTatiana TommasiYan TongAkihiko ToriiAntonio TorralbaLorenzo TorresaniAndrea TorselloTali TreibitzRudolph TriebelBill TriggsRoberto TronTomasz TrzcinskiIvor TsangYanghai TsinZhuowen TuTony TungPavan TuragaEngin TuretkenOncel TuzelGeorgios TzimiropoulosNorimichi UkitaMartin UrschlerArash VahdatJulien ValentinMichel ValstarKoen van de SandeJoost van de WeijerAnton van den HengelJan van GemertDaniel VaqueroKiran VaranasiMayank VatsaAshok VeeraraghavanOlga VekslerAlexander VezhnevetsRene VidalSudheendra

VijayanarasimhanJordi VitriaChristian VoglerCarl VondrickSven WachsmuthStefan WalkChaohui Wang

Jingdong WangJue WangRuiping WangKai WangLiang WangXinggang WangXin-Jing WangYang WangHeng WangYu-Chiang Frank WangSimon WarfieldYichen WeiYair WeissGordon WetzsteinOliver WhyteRichard WildesChristopher WilliamsLior WolfKwan-Yee Kenneth

WongOliver WoodfordJohn WrightChangchang WuXinxiao WuYing WuTianfu WuYang WuYingnian WuJonas WulffYu XiangTao XiangJianxiong XiaoDong XuLi XuYong XuKota YamaguchiTakayoshi YamashitaShuicheng YanJie YangQingxiong YangRuigang YangMeng YangYi YangChih-Yuan YangJimei Yang

XX Organization

Bangpeng YaoAngela YaoDit-Yan YeungAlper YilmazLijun YinXianghua YingKuk-Jin YoonShiqi YuStella YuJingyi YuJunsong YuanLu YuanAlan YuilleRamin ZabihChristopher Zach

Stefanos ZafeiriouHongbin ZhaLei ZhangJunping ZhangShaoting ZhangXiaoqin ZhangGuofeng ZhangTianzhu ZhangNing ZhangLei ZhangLi ZhangBin ZhaoGuoying ZhaoMing ZhaoYibiao Zhao

Weishi ZhengBo ZhengChangyin ZhouHuiyu ZhouKevin ZhouBolei ZhouFeng ZhouJun ZhuXiangxin ZhuHenning ZimmerKarel ZimmermannAndrew ZissermanLarry ZitnickDaniel Zoran

Additional Reviewers

Austin AbramsHanno AckermannDaniel AdlerMuhammed Zeshan

AfzalPulkit AgrawalEdilson de AguiarUnaiza AhsanAmit AidesZeynep AkataJon AlmazanDavid AltamarMarina AltermanMohamed Rabie AmerManuel AmthorShawn AndrewsOisin Mac AodhaFederica ArrigoniYuval BahatLuis BarriosJohn BastianFlorian BeckerC. Fabian

Benitez-QuirozVinay BettadapuraBrian G. Booth

Lukas BossardKatie BoumanHilton BristowDaniel CanelhasOlivier CanevetSpencer CappalloIvan Huerta CasadoDaniel CastroIshani ChakrabortyChenyi ChenSheng ChenXinlei ChenWei-Chen ChiuHang ChuYang CongSam Corbett-DaviesZhen CuiMaria A. DavilaOliver DemetzMeltem DemirkusChaitanya DesaiPengfei DouRalf DragonLiang DuDavid EigenJakob Engel

Victor EscorciaSandro EsquivelNicola FioraioMichael FirmanAlex FixOliver FleischmannMarco FornoniDavid FouheyVojtech FrancJorge Martinez G.Silvano GallianiPablo GarridoEfstratios GavvesTimnit GebruGeorgios GiannoulisClement GodardAnkur GuptaSaurabh GuptaAmirhossein HabibianDavid HafnerTom S.F. HainesVladimir HaltakovChristopher HamXufeng HanStefan HeberYacov Hel-Or

Organization XXI

David HeldBenjamin HellJan HellerAnton van den HengelRobert HenschelSteven HicksonMichael HirschJan HosangShell HuZhiwu HuangDaniel HuberAhmad HumayunCorneliu IlisescuZahra ImanThanapong IntharahPhillip IsolaHamid IzadiniaEdward JohnsJustin JohnsonAndreas JordtAnne JordtCijo JoseDaniel JungMeina KanBen KandelVasiliy KarasevAndrej KarpathyJan KautzChangil KimHyeongwoo KimRolf KoehlerDaniel KohlsdorfSvetlana KordumovaJonathan KrauseTill KroegerMalte KuhlmannIlja KuzborskijAlina KuznetsovaSam KwakPeihua LiMichael LamMaksim LapinGil LeviAviad LevisYan Li

Wenbin LiYin LiZhenyang LiPengpeng LiangJinna LieQiguang LiuTianliang LiuAlexander LoktyushinSteven LovegroveFeng LuJake LussierXutao LvLuca MagriBehrooz MahasseniAravindh MahendranSiddharth MahendranFrancesco MalapelleMateusz MalinowskiSantiago ManenTimo von MarcardRicardo Martin-BruallaIacopo MasiRoberto MeccaTomer MichaeliHengameh MirzaalianKylia MiskellIshan MisraJavier MontoyaRoozbeh MottaghiPanagiotis MoutafisOliver MuellerDaniel MunozRajitha NavarathnaJames NewlingMohamed OmranVicente OrdonezSobhan Naderi PariziOmkar ParkhiNovi PatriciaKuan-Chuan PengBojan PepikjFederico PerazziLoic PeterAlioscia PetrelliSebastian Polsterl

Alison PouchVittal PremanchandranJames PrittsLuis PuigJulian QuirogaVignesh RamanathanRene RanftlMohammad RastegariS. Hussain RazaMichael RealeMalcolm ReynoldsAlimoor RezaChristian RichardtMarko RistinBeatrice RossiRasmus RotheNasa RoufAnirban RoyFereshteh SadeghiZahra SadeghipoorFaraz SaedaarTanner SchmidtAnna SeninaLee SeverskyYachna SharmaChen ShenJaven ShiTomas SimonGautam SinghBrandon M. SmithShuran SongMohamed SouiaiSrinath SridharAbhilash SrikanthaMichael StollAparna TanejaLisa TangMoria TauJ. Rafael TenaRoberto ToldoManolis TsakirisDimitrios TzionasVladyslav UsenkoDanny VeikhermanFabio Viola

XXII Organization

Minh VoChristoph VogelSebastian VolzJacob WalkerLi WanChen WangJiang WangOliver WangPeng WangJan Dirk WegnerStephan WengerScott WorkmanChenglei Wu

Yuhang WuFan YangMark YatskarBulent YenerSerena YeungKwang M. YiGokhan YildirimRyo YonetaniStanislav YotovChong YouQuanzeng YouFisher YuPei Yu

Kaan YucerClausius ZelenkaXing ZhangXinhua ZhangYinda ZhangJiejie ZhuShengqi ZhuYingying ZhuYuke ZhuAndrew Ziegler

Table of Contents

Tracking and Activity Recognition

Visual Tracking by Sampling Tree-Structured Graphical Models . . . . . . . 1Seunghoon Hong and Bohyung Han

Tracking Interacting Objects Optimally Using Integer Programming . . . . 17Xinchao Wang, Engin Turetken, Francois Fleuret, and Pascal Fua

Learning Latent Constituents for Recognition of Group Activitiesin Video . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Borislav Antic and Bjorn Ommer

Recognition

Large-Scale Object Classification Using Label Relation Graphs . . . . . . . . 48Jia Deng, Nan Ding, Yangqing Jia, Andrea Frome, Kevin Murphy,Samy Bengio, Yuan Li, Hartmut Neven, and Hartwig Adam

30Hz Object Detection with DPM V5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65Mohammad Amin Sadeghi and David Forsyth

Knowing a Good HOG Filter when You See It:Efficient Selection of Filters for Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Ejaz Ahmed, Gregory Shakhnarovich, and Subhransu Maji

Linking People in Videos with “Their” Names Using CoreferenceResolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Vignesh Ramanathan, Armand Joulin, Percy Liang, and Li Fei-Fei

Poster Session 1

Optimal Essential Matrix Estimation via Inlier-Set Maximization . . . . . . 111Jiaolong Yang, Hongdong Li, and Yunde Jia

UPnP: An Optimal O(n) Solution to the Absolute Pose Problem withUniversal Applicability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

Laurent Kneip, Hongdong Li, and Yongduek Seo

3D Reconstruction of Dynamic Textures in Crowd Sourced Data . . . . . . . 143Dinghuang Ji, Enrique Dunn, and Jan-Michael Frahm

3D Interest Point Detection via Discriminative Learning . . . . . . . . . . . . . . 159Leizer Teran and Philippos Mordohai

XXIV Table of Contents

Pose Locality Constrained Representation for 3D Human PoseReconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

Xiaochuan Fan, Kang Zheng, Youjie Zhou, and Song Wang

Synchronization of Two Independently Moving Cameraswithout Feature Correspondences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189

Tiago Gaspar, Paulo Oliveira, and Paolo Favaro

Multi Focus Structured Light for Recovering Scene Shapeand Global Illumination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Supreeth Achar and Srinivasa G. Narasimhan

Coplanar Common Points in Non-centric Cameras . . . . . . . . . . . . . . . . . . . 220Wei Yang, Yu Ji, Jinwei Ye, S. Susan Young, and Jingyi Yu

SRA: Fast Removal of General Multipath for ToF Sensors . . . . . . . . . . . . . 234Daniel Freedman, Yoni Smolin, Eyal Krupka, Ido Leichter,and Mirko Schmidt

Sub-pixel Layout for Super-Resolution with Images in the OcticGroup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250

Boxin Shi, Hang Zhao, Moshe Ben-Ezra, Sai-Kit Yeung,Christy Fernandez-Cull, R. Hamilton Shepard,Christopher Barsi, and Ramesh Raskar

Simultaneous Feature and Dictionary Learning for Image Set BasedFace Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

Jiwen Lu, Gang Wang, Weihong Deng, and Pierre Moulin

Read My Lips: Continuous Signer Independent Weakly SupervisedViseme Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281

Oscar Koller, Hermann Ney, and Richard Bowden

Multilinear Wavelets: A Statistical Shape Space for Human Faces . . . . . . 297Alan Brunton, Timo Bolkart, and Stefanie Wuhrer

Distance Estimation of an Unknown Person from a Portrait . . . . . . . . . . . 313Xavier P. Burgos-Artizzu, Matteo Ruggero Ronchi,and Pietro Perona

Probabilistic Temporal Head Pose Estimation Using a HierarchicalGraphical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 328

Meltem Demirkus, Doina Precup, James J. Clark, and Tal Arbel

Description-Discrimination Collaborative Tracking . . . . . . . . . . . . . . . . . . . 345Dapeng Chen, Zejian Yuan, Gang Hua, Yang Wu,and Nanning Zheng

Online, Real-Time Tracking Using a Category-to-Individual Detector . . . 361David Hall and Pietro Perona

Table of Contents XXV

Robust Visual Tracking with Double Bounding Box Model . . . . . . . . . . . . 377Junseok Kwon, Junha Roh, Kyoung Mu Lee, and Luc Van Gool

Tractable and Reliable Registration of 2D Point Sets . . . . . . . . . . . . . . . . . 393Erik Ask, Olof Enqvist, Linus Svarm, Fredrik Kahl,and Giuseppe Lippolis

Graduated Consistency-Regularized Optimization for Multi-graphMatching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407

Junchi Yan, Yin Li, Wei Liu, Hongyuan Zha, Xiaokang Yang,and Stephen Mingyu Chu

Optical Flow Estimation with Channel Constancy . . . . . . . . . . . . . . . . . . . . 423Laura Sevilla-Lara, Deqing Sun, Erik G. Learned-Miller,and Michael J. Black

Non-local Total Generalized Variation for Optical Flow Estimation . . . . . 439Rene Ranftl, Kristian Bredies, and Thomas Pock

Learning Brightness Transfer Functions for the Joint Recoveryof Illumination Changes and Optical Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 455

Oliver Demetz, Michael Stoll, Sebastian Volz,Joachim Weickert, and Andres Bruhn

Hipster Wars: Discovering Elements of Fashion Styles . . . . . . . . . . . . . . . . . 472M. Hadi Kiapour, Kota Yamaguchi, Alexander C. Berg,and Tamara L. Berg

From Low-Cost Depth Sensors to CAD: Cross-Domain 3D ShapeRetrieval via Regression Tree Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489

Yan Wang, Jie Feng, Zhixiang Wu, Jun Wang, and Shih-Fu Chang

Fast and Accurate Texture Recognition with Multilayer Convolutionand Multifractal Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505

Hicham Badri, Hussein Yahia, and Khalid Daoudi

Learning to Rank 3D Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520Oncel Tuzel, Ming-Yu Liu, Yuichi Taguchi, and Arvind Raghunathan

Salient Color Names for Person Re-identification . . . . . . . . . . . . . . . . . . . . . 536Yang Yang, Jimei Yang, Junjie Yan, Shengcai Liao, Dong Yi,and Stan Z. Li

Learning Discriminative and Shareable Features for SceneClassification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 552

Zhen Zuo, Gang Wang, Bing Shuai, Lifan Zhao,Qingxiong Yang, and Xudong Jiang

XXVI Table of Contents

Image Retrieval and Ranking via Consistently ReconstructingMulti-attribute Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 569

Xiaochun Cao, Hua Zhang, Xiaojie Guo, Si Liu, and Xiaowu Chen

Neural Codes for Image Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584Artem Babenko, Anton Slesarev, Alexandr Chigorin, andVictor Lempitsky

Architectural Style Classification Using Multinomial Latent LogisticRegression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600

Zhe Xu, Dacheng Tao, Ya Zhang, Junjie Wu, and Ah Chung Tsoi

Instance Segmentation of Indoor Scenes Using a Coverage Loss . . . . . . . . 616Nathan Silberman, David Sontag, and Rob Fergus

Superpixel Graph Label Transfer with Learned Distance Metric . . . . . . . . 632Stephen Gould, Jiecheng Zhao, Xuming He, and Yuhang Zhang

Precision-Recall-Classification Evaluation Framework:Application to Depth Estimation on Single Images . . . . . . . . . . . . . . . . . . . 648

Guillem Palou Visa and Philippe Salembier

A Multi-stage Approach to Curve Extraction . . . . . . . . . . . . . . . . . . . . . . . . 663Yuliang Guo, Naman Kumar, Maruthi Narayanan,and Benjamin Kimia

Geometry Driven Semantic Labeling of Indoor Scenes . . . . . . . . . . . . . . . . 679Salman Hameed Khan, Mohammed Bennamoun, Ferdous Sohel,and Roberto Togneri

A Novel Topic-Level Random Walk Framework for Scene ImageCo-segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695

Zehuan Yuan, Tong Lu, and Palaiahnakote Shivakumara

Surface Matching and Registration by Landmark Curve-DrivenCanonical Quasiconformal Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 710

Wei Zeng and Yi-Jun Yang

Motion Words for Videos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 725Ekaterina H. Taralova, Fernando De la Torre, and Martial Hebert

Activity Group Localization by Modeling the Relations amongParticipants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 741

Lei Sun, Haizhou Ai, and Shihong Lao

Finding Coherent Motions and Semantic Regions in Crowd Scenes:A Diffusion and Clustering Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 756

Weiyue Wang, Weiyao Lin, Yuanzhe Chen, Jianxin Wu,Jingdong Wang, and Bin Sheng

Table of Contents XXVII

Semantic Aware Video Transcription Using Random ForestClassifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 772

Chen Sun and Ram Nevatia

Ranking Domain-Specific Highlights by Analyzing Edited Videos . . . . . . . 787Min Sun, Ali Farhadi, and Steve Seitz

A Multi-transformational Model for Background Subtractionwith Moving Cameras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 803

Daniya Zamalieva, Alper Yilmaz, and James W. Davis

Learning and Inference

Visualizing and Understanding Convolutional Networks . . . . . . . . . . . . . . . 818Matthew D. Zeiler and Rob Fergus

Part-Based R-CNNs for Fine-Grained Category Detection . . . . . . . . . . . . . 834Ning Zhang, Jeff Donahue, Ross Girshick, and Trevor Darrell

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 851