The conference seeks to provide a forum for presentation and discussion of original research papers on all aspects of
grammatical inference including, but not limited to:

  • Theoretical aspects of grammatical inference: learning paradigms, learnability results, complexity of learning.
  • Efficient learning algorithms for language classes inside and outside the Chomsky hierarchy. Learning tree and graph grammars. Learning distributions over strings, trees or graphs.
  • Theoretical and experimental analysis of different approaches to grammar induction, including artificial neural networks, statistical methods, symbolic methods, information-theoretic approaches, minimum description length, complexity-theoretic approaches, heuristic methods, etc.
  • Novel approaches to grammatical inference: Induction by DNA computing or quantum computing, evolutionary approaches, new representation spaces, etc.
  • Successful applications of grammatical inference to tasks in natural language processing, bioinformatics, machine translation, pattern recognition, language acquisition, software engineering, computational linguistics, spam and malware detection, cognitive psychology, etc.

Program Committee

  • Pieter Adriaans (Universiteit van Amsterdam, The Netherlands)
  • Dana Angluin (Yale University, USA)
  • Tom Armstrong (Wheaton College, USA)
  • Robert Berwick (MIT, USA)
  • John Case (University of Delaware, USA)
  • Alexander Clark (University of London, United Kingdom)
  • François Coste (INRIA Rennes, France)
  • Colin de la Higuera (Université de Nantes - LINA, France)
  • Henning Fernau (UniversitÀt Trier, Germany)
  • Pedro García (Universidad Politécnica de Valencia, Spain)
  • Jeffrey Heinz (University of Delaware, USA)
  • Makoto Kanazawa (National Institute of Informatics, Japan)
  • Satoshi Kobayashi (University of Electro-Communications, Japan)
  • Laurent Miclet (ENSSAT-Lannion, France)
  • Tim Oates (University of Maryland Baltimore County, USA)
  • Jose Oncina Carratala(Universidad de Alicante, Spain)
  • Yasubumi Sakakibara (Keio University, Japan)
  • Jose M. Sempere (Universidad Politécnica de Valencia, Spain)
  • Bernhard Steffen (University of Dortmund, Germany)
  • Etsuji Tomita (University of Electro-Communications, Japan)
  • Menno van Zaanen(Tilburg University, The Netherlands)
  • Sicco Verwer (Catholic University Leuven, Belgium)
  • Enrique Vidal (Universidad Politécnica de Valencia, Spain)
  • Ryo Yoshinaka (Hokkaido University, Japan)
  • François Yvon (Université Paris Sud 11, LIMSI, France)
  • Thomas Zeugmann (Hokkaido University, Japan)

The British Machine Vision Conference is one of the major international conferences on computer vision and related areas. Organized by the British Machine Vision Association, the 23rd BMVC was held at the University of Surrey, in the city of Guildford.

Welcome to BMVC 2012, it is our pleasure to host the conference at the University of Surrey in Guildford. This is the 23rd BMVC since its inception in 1990 and the second time Surrey has organised the conference. Almost 20 years have passed since BMVC in 1993 when it was chaired by John Illingworth with the support of the Centre for Vision Speech and Signal Processing (CVSSP) at the University of Surrey. We decided it was time for BMVC to return and our thanks go to our colleagues within CVSSP and to John for his contributions to the organisation. In fact BMVC goes back even further in the form of the Alvey Vision Club and our conference T-shirt gives a list of dates and venues for BMVC going back to the early Alvey days of the 80's.

BMVC has always maintained a single track and the format has changed little over the years. As the conference has grown in popularity its quality has increased, making it a prestigious event on the vision calendar. This year we received well over 400 submissions and accepted 8% of these for oral presentation and a further 24% for posters.

For 2012, we introduced changes to the reviewing process. We increased the reviewer pool to keep the loading low, despite increased submissions, and to ensure that reviewers could invest more time in their task. At least one reviewer per paper was nominated by an area chair. We also doubled the number of area chairs so that two area chairs were allocated to each paper. Unlike other conferences which operate a buddy system, our area chairs worked anonymously and independently and every paper had 2 recommendations justified in 2 consolidation reports. ACs who gave conflicting acceptance recommendations were invited into discussion to reach a consensus. As each AC had a different batch of papers, we were looking for consistency across both reviewers and area chairs to provide a robust decision making process. Oral papers were selected based on the reviews, AC consolidation reports, ranking in ACs. batches and suitability of the content for the general audience. Feedback indicates that this process has worked well and reviewers and area chairs from all over the world have worked hard to put together the highest quality program possible.

BMVC 2012 is a truly international conference. 67.5% of accepted papers are from Europe including 29% from the UK. There is also a significant 19.5% from North as well as from South America and 13% from Asia and Australia. The quality of research in these diverse locations is exceptionally high and it is also reflected in the selection of papers for oral presentations.

Sticking with the traditional single track format, BMVC starts on Monday afternoon with two tutorials; "Large-scale and larger-scale image search" by Dr Herve Jegou, INRIA and "MAP inference in Discrete Models" by Dr Pushmeet Kohli from Microsoft Research. We are also pleased to be able to welcome two invited speakers for the main conference; Prof Stan Sclaroff, Boston University (People in Motion: Pose, Action and Communication) and Prof. Jiri Matas, Czech Technical University (Visual Tracking in the 21st Century). On Tuesday evening there is a welcome reception hosted at Guildford Cathedral. While on Wednesday the conference banquet will be held at Brooklands Museum, the home of British Motorsport, and delegates will have exclusive access to the museum over drinks before dinner and awards in the clubhouse. On Friday, the BMVC Student Workshop will take place. This workshop has become a regular feature of BMVC and gives students in computer vision an opportunity to network and start collaborations at an early stage in their research career. The workshop will be single track containing both oral and poster presentations with keynote talks from Ariadna Quattoni UPC, Barcelona on .Latent Variable Models for Content-Based Image Retrieval and Structure Prediction. and Andrew Davison, Imperial College London on "Monocular SLAM and Real-Time Scene Perception".

As ever, BMVC proceedings are published by the BMVA as open access with the copyright retained by the authors. Extended abstracts of all the accepted contributions appear in this book. The full papers are published electronically on a USB key provided to delegates, and on the web with associated DOIs. Furthermore, the whole proceedings will be recorded and hosted online by and we are extremely grateful to the EU PASCAL2 network and video lectures for providing this service along with sponsorship of both the main conference and student workshop. We would also like to acknowledge the financial assistance of Microsoft Research, Stemmer Imaging and Toshiba all of which generously contributed to the conference as Gold sponsors. To 3DMD and Scorpion Vision as Silver sponsors and to Springer UK. We would like to thank the local organisation committee: Teofilo de Campos, John Illingworth and Fei Yan and especially Helen Cooper for taking great pleasure in doing all the organising we didn't want to do. We would also like to thank the student helpers and members of CVSSP who contributed, invited speakers and tutorial speakers and of course a huge debt to all the reviewers and area chairs for all the work they put in. Lastly many thanks to the authors for submitting the material to make this the best BMVC yet. We sincerely hope that you all have both a rewarding and enjoyable conference.


Richard Bowden, John Collomosse, Krystian Mikolajczyk
BMVC2012 General chairs

Lourdes Agapito Queen Mary, University of London
Abhir Bhalerao University of Warwick
Mathew Blaschko University of Oxford
Edmond Boyer INRIA Grenoble
Toby Breckon Cranfield University
Gabriel Brostow University College London
Gavin Brown University of Manchester
Matthew Brown University of Bath
Tilo Burghart University of Bristol
Andrea Cavallaro Queen Mary, University of London
Mike Chantler Heriot-Watt University
Ondrej Chum Czech Technical University Prague
Adrian Clark University of Essex
Tim Cootes University of Manchester
Tim Ellis Kingston University
Mark Everingham University of Leeds
Vittorio Ferrari University of Edinburgh
Andrew Fitzgibbon Microsoft Research Cambridge
Aphrodite Galata University of Manchester
Shaogang Gong Queen Mary, University of London
Lewis Griffin University College London (UCL)
Peter Hall University of Bath
Edwin Hancock University of York
Richard Harvey University of East Anglia
Adrian Hilton University of Surrey
Jesse Hoey University of Waterloo
David Hogg University of Leeds
Herve Jegou INRIA Rennes
Frederic Jurie University of Caen
Timor Kadir University of Oxford
Pushmeet Kohli Microsoft Research Cambridge
Xuelong Li Birkbeck, University of London
Derek Magee University of Leeds
Dave Marshall Cardiff University
Stephen Maybank Birkbeck, University of London
Walterio Mayol-Cuevas University of Bristol
Steve McKenna University of Dundee
Majid Mirmehdi University of Bristol
Florent Perronnin Xerox (XRCE) Grenoble
Stephen Pollard Hewlett Packard Labs UK
Tony Pridmore University of Nottingham
Ian Reid University of Oxford
Paul Rosin Cardiff University
Carsten Rother Microsoft Research Cambridge
Jon Starck The Foundry
Bjorn Stenger Toshiba Research Europe
Chris Taylor University of Manchester
Neil Thacker University of Manchester
Bill Triggs Laboratoire Jean Kuntzmann (LJK), Grenoble
Manuel Trucco University of Dundee
Andrea Vedaldi University of Oxford
Jakob Verbeek INRIA Grenoble
Kevin Wells University of Surrey
Tony Xiang Queen Mary, University of London,
Jianguo Zhang University of Dundee
Reyer Zwiggelaar Aberystwyth University


The Conference on Uncertainty in Artificial Intelligence (UAI) is the premier conference on issues relating to representation and management of uncertainty within the field of Artificial Intelligence. UAI is supported by the Association for Uncertainty in Artificial Intelligence (AUAI).



Uncertainty in Artificial Intelligence
Uncertainty in Artificial Intelligence

The fifteenth international conference on Artificial Intelligence and Statistics (AISTATS 2012) will be held on La Palma in the Canary Islands. AISTATS is an interdisciplinary gathering of researchers at the intersection of computer science, artificial intelligence, machine learning, statistics, and related areas. Since its inception in 1985, the primary goal of AISTATS has been to broaden research in these fields by promoting the exchange of ideas among them. We encourage the submission of all papers which are in keeping with this objective.

The 5th International Conference on Cognitive Systems brings together researchers in cognitive systems and robotics from academia, government and industry. The conference intends to present the state-of-the-art in cognitive systems and robotics in Europe and beyond. It aims to spotlight European research activities in the field and generate a lively discussion and interaction across disciplinary and institutional borders.

The International Conference on Pattern Recognition Applications and Methods would like to become a major point of contact between researchers, engineers and practitioners on the areas of Pattern Recognition, both from theoretical and application perspectives. Contributions describing applications of Pattern Recognition techniques to real-world problems, interdisciplinary research, experimental and/or theoretical studies yielding new insights that advance Pattern Recognition methods are especially encouraged.

The Twenty-fifth Annual Conference on Neural Information Processing Systems (NIPS) is a single-track machine learning and computational neuroscience conference that includes invited talks, demonstrations and oral and poster presentations of refereed papers. There are two satellite meetings: Women in Machine Learning and Casual Graphs.

NIPS 2011
NIPS 2011

The International Conference on Multimodal Interaction, ICMI 2011, will take place in Alicante (Spain), November 14-18, 2011, just after the ICCV 2011 (in Barcelona, Spain). ICMI is the fusion of the International Conference on Multimodal Interfaces and the Workshop on Machine Learning for Multimodal Interaction which, for the last two years held a combined event under the name ICMI-MLMI. Starting in this thirteenth edition the combined conference uses the new, shorter name.

ICMI is the premier international forum for multidisciplinary research on multimodal human-human and human-computer interaction, interfaces, and system development. The conference focuses on theoretical and empirical foundations, component technologies, and combined multimodal processing techniques that define the field of multimodal interaction analysis, interface design, and system development. ICMI 2011 will feature a single-track main conference which includes: keynote speakers, technical full and short papers (including oral and poster presentations), special sessions, demonstrations, exhibits and doctoral spotlight papers. The conference will be followed by workshops. The proceedings of ICMI 2011 will be published by ACM as part of their series of International Conference Proceedings.