4th UK Computer Vision Student Workshop (BMVW)
The BMVC Student Workshop will take place on Friday 7th September 2012, the day after the main BMVC conference, in the same venue. This workshop has become a regular feature of BMVC (see some of the previous editions: 2011, 2010). It 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.
Registration for this workshop is free for all UK students and BMVC registered participants.
Students studying in the UK are invited to submit full-length high-quality papers of which the main author is a student. All papers will be reviewed and selected for either oral or poster presentation. All accepted papers for the workshop will be digitally published in the USB stick provided at the conference and on-line at the BMVA website (e.g. follow this link for last year's proceedings).
|13 July||Paper submissions due|
|27 July||Author notifications|
|4 August||Camera ready papers due|
|20 August||Impromptu poster contributions due|
|7 September 2012||Workshop, from 9am to 4pm|
Please follow this link for the full workshop programme.
Latent Variable Models for Content-Based Image Retrieval and Structure Prediction
In the first part of the talk I will present recent work on learning latent variable models for content-based image retrieval. To learn a function that predicts the relevance of a database image to an image query all that we need is some form of feedback from users of the retrieval system. For example, we can obtain triplet constraints specifying that relative to some query Q, an image A should be ranked higher than an image B. When such feedback is available ranking SVMs can be used to induce the retrieval function. I will describe an extension of this framework where instead of learning a single relevance function we learn a mixture of relevance functions. Intuitively, given a query we first compute a distribution over "coarse" latent classes and then compute the relevance function for queries of that class. I will present a simple learning algorithm that induces both the latent classes and the parameters of each model.
In the second part of the talk I will describe some of my current work on developing efficient learning algorithms for structure prediction with latent variables. These algorithms are based on using an algebraic representation that exploits directly the markovianity of the distribution.
Imperial College London
Monocular SLAM and Real-Time Scene Perception
We have seen great advances in real-time 3D vision in recent years, enabled by algorithmic improvements, the continuing increase in commodity processing power and better camera technology. Research in Monocular SLAM (Simultaneous Localisation and Mapping), where a single agile camera moves through a mostly static scene, was for a long time focused on mapping only enough of a scene to enable robust real-time motion estimation of the camera itself. Attention is now turning however to gradually improving the quality of scene reconstruction which can be achieved in real-time. I will speak about how early work on feature-based SLAM is now being surpassed by methods which aim to map dense scene structure, and how this is leading towards ever-more general 3D scene modelling and understanding.
As with the main conference, the workshop topics include, but are not limited to:
- Statistics and machine learning for vision
- Stereo, calibration, geometric modelling and processing
- Person, face and gesture tracking
- Object and activity recognition
- Motion, flow and tracking
- Segmentation and feature extraction
- Model-based vision
- Image processing techniques and methods
- Texture, shape and colour
- Video analysis
- Document processing and recognition
- Vision for quality assurance, medical diagnosis, etc.
- Vision for visualization, interaction, and graphics
|Workshop chair:||Dr. Teo de Campos|
|Mark Barnard||University of Surrey|
|Barbara Caputo||IDIAP Research Institute|
|Alessio Del Bue||Istituto Italiano di Tecnologia|
|Carl Henrik Ek||KTH Stockholm|
|Nazli FarajiDavar||University of Surrey|
|Albert Gordo||Universidad Autonoma de Barcelona|
|Ashish Gupta||University of Surrey|
|Rui Hu||University of Surrey|
|Olaf Kahler||University of Oxford|
|Jan Knopp||KU Leuven|
|Piotr Koniusz||University of Surrey|
|Luca Marchesotti||Xerox Research Centre Europe|
|Alexander Mansfield||ETH Zurich|
|Julian McAuley||University of Stanford|
|Mukta Prasad||ETH Zurich|
|Erik Rodner||Friedrich Schiller University of Jena|
|Jose Rodriguez-Serrano||Xerox Research Centre Europe|
|Violet Snell||University of Surrey|
|Eric Sommerlade||University of Oxford|
|Phil Tresadern||University of Manchester|
|Ruixuan Wang||University of Dundee|
|Fei Yan||University of Surrey|
|Huiyu Zhou||Queen's University Belfast|
Instructions for Authors
New call for abstracts: please follow this link for an additional call for contributions to be presented as posters at the workshop.
Call for papers
This call has been closed.
We acknowledge authors for the quality of the submissions and reviewers for their thorough work. The text below will remain here for archiving purposes.
Submissions about ongoing work are welcome, but novelty is required. By submitting a manuscript to the workshop, the authors guarantee that it has not been published previously (nor accepted for publication) in substantially similar form. Violation of any of these conditions will lead to rejection. We also require that the first author of the paper be a student at a British institution.
Submissions of papers are handled by CMT: https://cmt.research.microsoft.com/BMVW2012/
The format of the papers is the same as for the main BMVC 2012 conference.
The length should not exceed 9 pages excluding the author list and references (i.e. measured from the top of the abstract to the top of the reference list.
Typesetting should preferably be done using the PDFLaTeX system (part of all modern LaTeX distributions). However, submissions prepared in Microsoft Word or OpenOffice will be accepted and should be submitted for review in PDF. Please use the templates provided.
Authors can optionally submit supplementary material limited to 10MB or cite the URL of their own webpages for downloading of larger files. Supplementary material may include:
- videos to showcase results/demo of the proposed approach/system,
- images and other results in addition to the ones in the paper,
- related submissions to conferences and journals,
- appendices or technical reports containing extended proofs and mathematical derivations that are not essential to the understanding of the submitted paper.
Videos should preferably be encoded using an MP4 codec such as DivX. Please include a README text file with each video specifying the exact codec used and a URL where the codec can be downloaded.
The authors should refer to the contents of the supplementary material appropriately in the paper.
For further information please contact the Workshop Chair, Dr Teo de Campos.