BMVC 2012 Conference Programme


The list of papers accepted to BMVC 2012 is now available.



Time

Monday - 3th September

12:30 Registration desk opens
13:30 Tutorial 1. Herve Jegou  -  Large scale and larger-scale image search
15:30 Coffee
16:00 Tutorial 2. Pushmeet Kohli  -  MAP inference in Discrete Models.




Time

Tuesday - 4th September

8:30 Registration desk opens & Coffee
9:00 Conference welcome
9:10 Keynote 1. Stan Sclaroff  -  People In Motion: Pose, Action and Communication
10:10 Session 1: Tracking
Session chair: Richard Bowden
  • Automatic and Efficient Long Term Arm and Hand Tracking for Continuous Sign Language TV Broadcasts
    Tomas Pfister (University of Oxford), James Charles (University of Leeds), Mark Everingham (University of Leeds), Andrew Zisserman (University of Oxford)
  • Deformable Tracking of Textured Curvilinear Objects
    Nicolas Padoy (Johns Hopkins University), Gregory Hager (Johns Hopkins University)
10:50 Coffee
11:20 Session 2: People and Pose
Session chair: Adrian Hilton
  • Using Richer Models for Articulated Pose Estimation of Footballers
    Vahid Kazemi (KTH, Sweden), Josephine Sullivan (KTH, Sweden)
  • Dynamical Pose Filtering for Mixtures of Gaussian Processes
    Martin Fergie (Universiy of Manchester), Aphrodite Galata (University of Manchester)
  • Close-Range Human Detection and Tracking for Head-Mounted Cameras
    Dennis Mitzel (RWTH Aachen University), Bastian Leibe (RWTH Aachen University)
  • Detection and Tracking of Occluded People
    Siyu Tang (Max Planck Institute for Informatics), Mykhaylo Andriluka (Max Planck Institute for Informatics), Bernt Schiele (Max Planck Institute for Informatics)
  • Latent SVMs for Human Detection with a Locally Affine Deformation Field
    L’ubor Ladický (University of Oxford), Philip Torr (Oxford Brookes University), Andrew Zisserman (University of Oxford)
13:00 Lunch
14:00 Poster Session 1
15:00 Tea
15:30 Session 3: Segmentation
Session chair: Toby Breckon
  • Stixmentation - Probabilistic Stixel based Traffic Scene Labeling
    Friedrich Erbs (Daimler AG), Beate Schwarz (Daimler AG), Uwe Franke (Daimler AG)
  • MoT - Mixture of Trees Probabilistic Graphical Model for Video Segmentation
    Ignas Budvytis (University of Cambridge), Vijay Badrinarayanan (University of Cambridge), Roberto Cipolla (University of Cambridge)
  • Improved Initialization and Gaussian Mixture Pairwise Terms for Dense Random Fields with Mean-field Inference
    Vibhav Vineet (Oxford Brookes University), Jonathan Warrell (Oxford Brookes University), Paul Sturgess (Oxford Brookes Vision Group), Philip Torr (Oxford Brookes University)
  • Image Segmentation using Dual Distribution Matching
    Tatsunori Taniai (University of Tokyo), Viet-quoc Pham (Toshiba Corporation, Japan), Keita Takahashi (University of Electro-Communications, Japan), Takeshi Naemura (University of Tokyo)
16:50 BMVA Annual General Meeting (AGM)
19:00 Conference Reception (Guildford Cathedral)




Time

Wednesday - 5th September

8:30 Registration desk opens & Coffee
9:00 Keynote 2. Jiri Matas  -  Visual Tracking in the 21st Century
10:00 Session 4: Retrieval
Session chair: John Collomosse
  • Image Retrieval for Image-Based Localization Revisited
    Torsten Sattler (RWTH Aachen University), Tobias Weyand (RWTH Aachen University), Bastian Leibe (RWTH Aachen University), Leif Kobbelt (RWTH Aachen University)
  • Improved Geometric Verification for Large Scale Landmark Image Collections
    Rahul Raguram (University of North Carolina at Chapel Hill), Joseph Tighe (University of North Carolina at Chapel Hill), Jan-michael Frahm (University of North Carolina at Chapel Hill)
10:40 Coffee
11:10 Session 5: Object Recognition
Session chair: Andrea Vedaldi
  • Transfer Learning by Ranking for Weakly Supervised Object Annotation
    Zhiyuan Shi (Queen Mary, University of London), Parthipan Siva (Queen Mary, University of London), Tao Xiang (Queen Mary, University of London)
  • Enhancing Exemplar SVMs using Part Level Transfer Regularization
    Yusuf Aytar (University of Oxford), Andrew Zisserman (University of Oxford)
  • Do We Need More Training Data or Better Models for Object Detection?
    Xiangxin Zhu (University of California), Carl Vondrick (MIT), Deva Ramanan (University of California), Charless Fowlkes (University of California)
  • An Object Co-occurrence Assisted Hierarchical Model for Scene Understanding
    Xin Li (Temple University, Philadelphia), Yuhong Guo (Temple University, Philadelphia)
  • Efficient Kernels Couple Visual Words Through Categorical Opponency
    Ioannis Alexiou (Imperial College London), Anil Bharath (Imperial College London)
12:50 Lunch
14:00 Poster Session 2, including Demos & Exhibition
15:00 Tea
15:30 Session 6: Action Recognition
Session chair: Charless Fowlkes
  • Learning discriminative space-time actions from weakly labelled videos
    Michael Sapienza (Oxford Brookes University), Fabio Cuzzolin (Oxford Brookes University), Philip Torr (Oxford Brookes University)
  • Spatio-Temporal Convolutional Sparse Auto-Encoder for Sequence Classification
    Moez Baccouche (Orange Labs R&D), Franck Mamalet (Orange Labs R&D), Christian Wolf (LIRIS laboratory - INSA Lyon), Christophe Garcia (LIRIS laboratory - INSA Lyon), Atilla Baskurt (LIRIS laboratory - INSA Lyon)
  • Cross-View Action Recognition via a Transferable Dictionary Pair
    Jingjing Zheng (University of Maryland), Zhuolin Jiang (University of Maryland), Jonathon Phillips (National Institute of Standards and Technology), Rama Chellappa (University of Maryland)
  • A Videography Analysis Framework for Video Retrieval and Summarization
    Kang Li (SUNY at Buffalo), Sangmin Oh (Kitware Inc), A.g. Amitha Perera (Kitware Inc), Yun Fu (Northeastern University, Boston)
18:00 Conference Banquet (Brooklands Museum)




Time

Thursday - 6th September

8:30 Registration desk opens & Coffee
9:00 Session 7: Recognition and Face
Session chair: Gabriel Brostow
  • Scene Text Recognition using Higher Order Language Priors
    Anand Mishra (IIIT Hyderabad, India), Karteek Alahari (INRIA - WILLOW / ENS), Cv Jawahar (IIIT Hyderabad, India)
  • Data-Driven Scene Understanding from 3D Models
    Scott Satkin (The Robotics Institute, Carnegie Mellon University), Jason Lin (The Robotics Institute, Carnegie Mellon University), Martial Hebert (The Robotics Institute, Carnegie Mellon University)
  • Tom-vs-Pete Classifiers and Identity-Preserving Alignment for Face Verification
    Thomas Berg (Columbia University), Peter Belhumeur (Columbia University)
  • Let the Shape Speak - Discriminative Face Alignment using Conjugate Priors
    Pedro Martins (Institute of Systems Robotics, University of Coimbra), Rui Caseiro (Institute of Systems Robotics, University of Coimbra), João Henriques (Institute of Systems Robotics, University of Coimbra), Jorge Batista (Institute of Systems Robotics, University of Coimbra)
  • Dense Active Appearance Models Using a Bounded Diameter Minimum Spanning Tree
    Robert Anderson (University of Cambridge), Bjorn Stenger (Toshiba Research Europe), Roberto Cipolla (University of Cambridge)
10:40 Coffee
11:10 Session 8: Registration and Image Processing
Session chair: Walterio Mayol
  • PMBP: PatchMatch Belief Propagation for Correspondence Field Estimation
    Frederic Besse (University College London), Carsten Rother (Microsoft Research Cambridge), Andrew Fitzgibbon (Microsoft Research Cambridge), Jan Kautz (University College London)
  • Deformable 3D Reconstruction with an Object Database
    Pablo Alcantarilla (Université d'Auvergne), Adrien Bartoli (Université d'Auvergne)
  • Incremental Light Bundle Adjustment
    Vadim Indelman (Georgia Institute of Technology), Richard Roberts (Georgia Institute of Technology), Chris Beall (Georgia Institute of Technology), Frank Dellaert (Georgia Institute of Technology)
  • Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding
    Marco Bevilacqua (INRIA), Aline Roumy (INRIA), Christine Guillemot (INRIA), Marie-line Alberi Morel (Bell Labs France)
  • Fast Non-uniform Deblurring using Constrained Camera Pose Subspace
    Zhe Hu ( University of California at Merced), Ming-hsuan Yang (University of California at Merced)
13:10 Closing Remarks


Friday - 7th September

The 4th UK Computer Vision Student Workshop (BMVW) immediately follows on Friday.