ECCV_005.pdf

所需积分/C币:9 2018-07-13 09:41:33 178.07MB PDF

ECCV (European Conference on Computer Vision) ECCV_005.pdf
Moreinformationaboutthisseriesathttp://www.springer.com/series/7412 Bastian leibe. Jiri matas Nicu Sebe. Max Welling(Eds) Computer vision ECCV 2016 14th European Conference Amsterdam. The Netherlands. October 11-14.2016 Proceedings, Part V pringer editors Bastian leibe Nicu sebe RWTH Aachen University of Trento Aachen P Germany Ital Jiri matas Max welling Czech Technical University University of Amsterdam Prague Amsterdam Czech Republic The netherland ISSN03029743 Issn 1611-3349(electronic) Lecture Notes in Computer Science ISBN978-3-319-46453-4 ISBN978-3-31946454-1( eBook) DOI10.1007/978-3-31946454-1 Library of Congress Control Number: 2016951693 LNCS Sublibrary: SL6-Image Processing, Computer Vision, Pattern Recognition, and Graphics C Springer International Publishing AG 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the 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 information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made Printed on acid-free pap This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse ll, 6330 Cham, Switzerland Foreword Welcome to the proceedings of the 2016 edition of the European Conference on Computer Vision held in Amsterdam! It is safe to say that the european Conference on Computer vision is one of the top conferences in computer vision. It is good to reiterate the history of the conference to see the broad base the conference has built in its 13 editions. First held in 1990 in Antibes(France), it was followed by subsequent con ferences in Santa Margherita Ligure (Italy) in 1992, Stockholm (Sweden)in 1994, Cambridge (UK)in 1996, Freiburg( Germany) in 1998, Dublin (Ireland) in 2000, Copenhagen(Denmark) in 2002, Prague(Czech Republic)in 2004, Graz(Austria)in 2006, Marseille(france)in 2008, Heraklion( Greece)in 2010, Florence (Italy)in 2012 and zurich(Switzerland)in 2014 For the 14th edition, many people worked hard to provide attendees with a most warm welcome while enjoying the best science. The Program Committee, Bastian eibe, Jiri Matas, Nicu Sebe, and Max Welling, did an excellent job. Apart from the scientific program, the workshops were selected and handled by Herve Jegou and Gang Hua, and the tutorials by Jacob Verbeek and Rita Cucchiara. Thanks for the great job The coordination with the subsequent ACM Multimedia offered an opportunity to expand the tutorials with an additional invited session, offered by the University of Amsterdam and organized together with the help of ACM Multimedia Of the many people who worked hard as local organizers we would like to single out Martine de wit of the Uva Conference Office, who delicately and efficiently organized the main body. also the local organizers hamdi Dibeklioglu, Efstratios Gavves, Jan van Gemert Thomas mensink. and Mihir jain had their hands full. as a venue. we chose the Royal Theatre Carre located on the canals of the amstel river in downtown amsterdam Space in Amsterdam is sparse, so it was a little tighter than usual. The university lent us their downtown campuses for the tutorials and the workshops. a relatively new thing was the industry and the sponsors for which ronald poppe and peter de with did a great job, while Andy Bagdanov and John Schavemaker arranged the demos. Michael Wilkinson took care to make Yom Kippur as comfortable as possible for those for whom it is an important day. We thank Marc Pollefeys, Alberto del Bimbo, and virginie mes for their advice and help behind the scenes We thank all the anonymous volunteers for their hard and precise work. We also thank our generous sponsors. Their support is an essential part of the program. It is good to see such a level of industrial interest in what our community is doing Amsterdam does not need any introduction. Please emerge yourself but do not drown in it have a nice time October 2016 Theo ge Arnold smeulders Preface Welcome to the proceedings of the 2016 European Conference on Computer vision (ECCV 2016) held in Amsterdam, The Netherlands. We are delighted to present this volume reflecting a strong and exciting program, the result of an extensive review process. In total, we received 1, 561 paper submissions. Of these, 81 violated the eCcv submission guidelines or did not pass the plagiarism test and were rejected without reviewWeemployedtheithENticatesoftware(www.ithenticate.com)forplagiarism detection Of the remaining papers, 415 were accepted (26.6 %): 342 as posters(22.6%), 45 as spotlights (2.9 %), and 28 as oral presentations(1. 8 %). The spotlights -short, five- minute podium presentations are novel to eccv and were introduced after their success at the CVPr 2016 conference. All orals and spotlights are presented as posters as well. The selection process was a combined effort of four program co-chairs(PCs) 74 area chairs(ACs), 1,086 Program Committee members, and 77 additional reviewers As PCs, we were primarily responsible for the design and execution of the review process. Beyond administrative rejections, we were involved in acceptance decisions only in the very few cases where the aCs were not able to agree on a decision PCs, as is customary in the field, were not allowed to co-author a submission general co-chairs and other co-organizers played no role in the review process, were permitted to submit papers, and were treated as any other author. Acceptance decisions were made by two independent ACs. There were 74 ACs selected by the PCs according to their technical expertise, experience, and geographical diversity(41 from European, five from Asian, two from Australian, and 26 from North American institutions ). The ACs were aided by 1,086 Program Committee members to whom papers were assigned for reviewing. There were 77 additional reviewers, each supervised by a Program Committee member. The Program Committee was selected from committees of previous eccv, iccv, and CVPr conferences and was extended on the basis of suggestions from the aCs and the PCs. Having a large pool of Progra Committee members for reviewing allowed us to match expertise while bounding reviewer loads. Typically five papers, but never more than eight, were assigned to a Program Committee member. Graduate students had a maximum of four papers to review The eCCv 2016 review process was in principle double-blind. Authors did not know reviewer identities, nor the ACs handling their paper(s). However, anonymity becomes difficult to maintain as more and more submissions appear concurrently on arXiv. org This was not against the eccv 2016 double submission rules which followed the practice of other major computer vision conferences in the recent past. The existence of ar Xiv publications, mostly not peer-reviewed, raises difficult problems with the assessment of unpublished, concurrent, and prior art, content overlap, plagiarism, and self-plagiarism. Moreover, it undermines the anonymity of submissions We found that not all cases can be covered by a simple set of rules. Almost all controversies during the review process were related to the ar Xiv issue. Most of the reviewer inquiries were VIII P trace resolved by giving the benefit of the doubt to eCcv authors. However, the problem will have to be discussed by the community so that consensus is found on how to handle the issues brought by publishing on arXiv Particular attention was paid to handling conficts of interest. Conflicts of interest between ACs, Program Committee members, and papers were identified based on the authorship of ECC V 2016 submissions, on the home institutions, and on previous col- laborations of all researchers involved To find institutional conficts all authors program Committee members and acs were asked to list the internet domains of their current institutionsTofindcollaboratorsthereSearcherccdatabase(http://researcher.cc/) funded by the Computer Vision Foundation, was used to find any co-authored papers in the period 2012-2016. We pre-assigned approximately 100 papers to each AC, based on affinity scores from the Toronto Paper matching System. ACs then bid on these indi cating their level of expertise. Based on these bids, and conflicts of interest, approxi mately 40 papers were assigned to each ac. The aCs then suggested seven reviewers from the pool of program Committee members for each paper, in ranked order, from which three were chosen automatically by CMT ( Microsofts Academic Conference Management Service), taking load balancing and conficts of interest into account The initial reviewing period was five weeks long, after which reviewers provided reviews with preliminary recommendations. With the generous help of several last minute reviewers, each paper received three reviews. Submissions with all three reviews suggesting rejection were independently checked by two ACs and if they agreed, the manuscript was rejected at this stage (early rejects"). In total, 334 manuscripts(22.5 %) were early-rejected, reducing the average AC load to about 30 Authors of the remaining submissions were then given the opportunity to rebut the reviews, primarily to identify factual errors. Following this, reviewers and ACs dis cussed papers at length, after which reviewers finalized their reviews and gave a final recommendation to the ACs. Each manuscript was evaluated independently by two ACs who were not aware of each others identities. In most of the cases after extensive discussions, the two ACs arrived at a common decision, which was always adhered to by the PCs. In the very few borderline cases where an agreement was not reached, the PCS acted as tie-breakers. Owing to the rapid expansion of the field which led to an unexpectedly large increase in the number of submissions the size of the venue became a limiting factor and a hard upper bound on the number of accepted papers had to be imposed. We were able to increase the limit by replacing one oral session by a poster session. Nevertheless, this forced the PCs to reject some borderline papers that could otherwise have been accepted We want to thank everyone involved in making the ECC V 2016 possible. First and foremost, the success of ECC V 2016 depended on the quality of papers submitted by the authors, and on the very hard work of the ACs, the Program Committee members, and the additional reviewers we are particularly grateful to rene vidal for his continuous support and sharing experience from organizing ICCV 2015, to Laurent Charlin for the use of the toronto Paper Matching System, to Ari Kobren for the use of the researcher. cc tools, to the Computer Vision Foundation(CvF) for facilitating the use of the iThenticate plagiarism detection software, and to Gloria Zen and radu - Laurentiu Vieriu for setting up CMT and managing the various tools involved. We also owe a debt of gratitude for the support of the Amsterdam local organizers, especially Hamdi Dibeklioglu for keeping the Pr eface IX website always up to date finally the preparation of these proceedings would not have been possible without the diligent effort of the publication chairs, Albert Ali salah and Robby tan, and of Anna Kramer from Springer. October 2016 Bastian leibe Jiri matas Nicu sebe Max Welling Organization General chairs Theo gevers University of Amsterdam, The Netherlands Arnold smeulders University of Amsterdam, The Netherlands Program Committee Co-chairs Bastian leibe RWTH Aachen, Germany Jiri matas Czech Technical University, Czech Republic Nicu sebe University of Trento. Ital Max welling University of Amsterdam, The Netherlands Honorary Chair Jan Koenderink Delft University of Technology, The Netherlands and Ku leuven belgium Advisory program Chair Luc van gool EtH Zurich. switzerland Advisory Workshop Chair Josef Kittler University of Surrey, UK Advisory Conference Chair alberto del bimbo University of florence, Italy Local Arrangements Chairs Hamdi dibeklioglu Delft University of Technology, The Netherlands Efstratios gavves University of amsterdam, The Netherlands Jan van gemert Delft University of Technology, The Netherlan Thomas mensink University of Amsterdam, The Netherlands Michael wilkinson University of Groningen, The Netherlands

...展开详情
试读 127P ECCV_005.pdf

评论 下载该资源后可以进行评论 1

singular2611 不错不错不错
2019-04-14
回复
img
三七开开
  • GitHub

    绑定GitHub第三方账户获取

关注 私信 TA的资源

上传资源赚积分,得勋章
    最新推荐
    ECCV_005.pdf 9积分/C币 立即下载
    1/127
    ECCV_005.pdf第1页
    ECCV_005.pdf第2页
    ECCV_005.pdf第3页
    ECCV_005.pdf第4页
    ECCV_005.pdf第5页
    ECCV_005.pdf第6页
    ECCV_005.pdf第7页
    ECCV_005.pdf第8页
    ECCV_005.pdf第9页
    ECCV_005.pdf第10页
    ECCV_005.pdf第11页
    ECCV_005.pdf第12页
    ECCV_005.pdf第13页
    ECCV_005.pdf第14页
    ECCV_005.pdf第15页
    ECCV_005.pdf第16页
    ECCV_005.pdf第17页
    ECCV_005.pdf第18页
    ECCV_005.pdf第19页
    ECCV_005.pdf第20页

    试读已结束,剩余107页未读...

    9积分/C币 立即下载 >