Electronic Proceedings of the
5th UK Computer Vision Student Workshop (BMVW)

Editor: Jose Martinez-Carranza

 
BMVC2013 Student Workshop Logo

 


Workshop Foreword

 

Since its first edition in 2009, the BMVC Student workshop has become a venue where postgraduate students of British institutions can present original results of their research. It also offers the opportunity to meet and network with other students, thus opening the avenue for future collaborations.

As in previous years, a call for papers was opened. Papers had to follow the same format of the main conference papers. Each paper was reviewed by two reviewers and in those cases where the decision was undecided, a meta-reviewer took over in order to decide the finale outcome.

This year four papers were accepted to be presented orally in the workshop. Each one of these papers has been published in the Proceedings of the 5th UK Student Workshop on Computer Vision - BMVW 2013. Two papers where distinguished during the closing remarks of the workshop. One with Honour Mention and one with the Best Workshop Paper Award. These papers are indicated in the proceedings below.

Furthermore, students of British institutions whose work has been accepted for publication in the main conference (BMVC) but whose presentation format is that of a poster, have the opportunity of presenting their work orally in the workshop. This year 4 students kindly accepted the invitation to present their work orally in the workshop. These talks were presented in the session 2 of the workshop (see programme at http://bmvc2013.bristol.ac.uk/workshopprogramme) and the corresponding papers can be found in the proceedings of the BMVC 2013. In addition to the presentations delivered by the student, the workshop also included two keynote sessions presented by two brilliant experts in their field: Gabriel Brostow, and Edward Rosten.

I take this opportunity to thank to this year's BMVC committee for having given me the opportunity to organise and host the student workshop and also for providing the necessary funding which made possible to invite two remarkable researchers to deliver the keynotes and also to cover the workshop expenses. I want to thank in particular to Tilo Burghardt whose enthusiasm gave me the confidence to organise the workshop, to Dima Damen and Majid Mirmehdi for their valuable support and advice, to Walterio Mayol whose support played a key role in obtaining funds to sponsor the workshop and to Antonina Timofejeva for her support on the admin details and for her help in the registration desk.

I am also grateful to Teo de Campos, last year's workshop chair, who shared his experience without reservations and gave me invaluable hints and advice on the organisation of the workshop. I am also thankful to Gabriel Brostow and Edward Rosten, who kindly accepted to present the keynotes and to the students who accepted the invitation to present their work in the workshop. Many thanks also go to the reviewers whose names are listed below.

Last but not least, I am grateful to Adeline Paiement and Osian Haines who wonderfully assisted me and who also helped during the reviewing process. I should also mention that Adeline designed the workshop's logo and Osian took wonderful pictures of the event. Thanks also to Anthony Glynn and Alan Torres who also helped on the day of the workshop.


Jose Martinez-Carranza
Workshop Chair

 


 

Reviewers

Andrew Gee Qualcommm
Carlos Arteta University of Oxford
Osian Haines Univeristy of Bristol
David Hanwell University of Bristol
Javier Orozco Imperial College London
Rob Frampton Univeristy of Bristol
Jack Greenhalgh University of Bristol
Lucilio Cordero-Grande University of Valladolid
Mark Hansen University West of England
Pablo Alcantarilla Georgia Institute of Technology
Philip Tresadern University of Manchester
Rui Hu University of Surrey
Violet Snell University of Surrey
Wenbin Li University of Bath
Adeline Paiement University of Bristol

 


 

Proceedings

Keynote 1
1.  If only we had tracked something like this before
Gabriel Browstow (UCL)
Session 1
5. Improving photometric stereo through per-pixel light vector calculation
Jahanzeb Ahmad, Jiuai Sun, Lyndon Smith, Melvyn Smith (University West of England)
2. Multi-Scale Azimuthal Projection Distance Image of Normal Maps for 3D Facial Skin Texture Analysis
Alassane Seck, Hannah Dee,Bernard Tiddman (Aberystwyth University)
Keynote 2
4.  Optimized Corner and Object Detection: a Completely Non Unified Approach
Edward Rosten (Computer Vision Consulting)
Session 2
3. Synthesising unseen image conditions to enhance classification accuracy for sparse datasets: applied to chimpanzee face recognition
Roz Sandwell, Alexander Loos, Tilo Burghardt (University of Bristol)
Honour Mention
6. A Multiple Motion Model Tracker Handling Occlusion and Rapid Motion Variation
Muhammad Haris Khan, Michel Valstar, Tony Pridmore (University of Nottingham)
Best Paper Award