Final Program

rigth click and save target as for program PDF document

Invited Talks

Technical Program

Opening

8:50    

Invited Talk 1

9:00  

Distributed Saliency Computations Solve the Feature Binding Problem
John K. Tsotsos
Department of Computer Science and Centre for Vision Research
York University, Toronto, Canada

Computational vision has a long history of proposing methods for decomposing a visual signal into its components. For example, many good strategies have appeared for decomposing visual motion signals. What has been far more elusive is how to recombine those components into a whole. This problem has even merited its own name - the binding problem. To date no realizable process has appeared to solve the binding problem, even in part, although several proposals are being studied. This presentation will focus on a new strategy utilizing a novel distributed saliency computation mechanism that solves at least one aspect of the binding problem, namely the binding of features from separate representations into a whole. Several examples will be drawn from a new, biologically realistic, motion analysis system, one that attends to complex motion patterns. An example of how this approach even yields Treisman-style illusory conjunctions is included. The entire process is implemented and operates on real image sequences. The implications for the neurobiology of visual attention will round out the presentation.

Session 1: Attention in Object and Scene Recognition

9:45   Object Based Visual Attention: Searching for Objects Defined by Size
Ola Ramstrom and Henrik I. Christensen
Royal Institute of Technology, Sweden
    Inherent Limitations of Visual Search and the Role of Inner-Scene Similarity
Tamar Avraham and Michael Lindenbaum
Technion Haifa, Israel
10:45   coffee break

Session 2: Architectures for Sequential Attention

11:00   Selective Attention for Identification Model (SAIM): Simulating Different Types of Visual Neglect
Dietmar Heinke and Glyn W. Humphreys
University of Birmingham, UK
    A Model of Object-Based Attention That Guides Active Visual Search to Behaviourally Relevant Locations
Linda J. Lanyon and Susan L Denham
University of Plymouth, UK
    Learning of Position and Attention-Shift Invariant Recognition across Attention Shifts
Muhua Li and James J. Clark
McGill University, Canada
12:15   lunch

Invited Talk 2

13:45  

The Computational Neuroscience of Visual Cognition: Attention, Memory and Reward
Gustavo Deco
Institucio Catalana de Recerca i Estudis Avancats (ICREA)
Universitat Pompeu Fabra, Barcelona, Spain

Cognitive behaviour requires complex context-dependent processing of information that emerges from the links between attentional perceptual processes, working memory and reward-based evaluation of the performed actions. We describe a computational neuroscience theoretical framework which shows how an attentional state held in a short term memory in the prefrontal cortex can by top-down processing influence ventral and dorsal stream cortical areas using biased competition to account for many aspects of visual attention. We also show how within the prefrontal cortex an attentional bias can influence the mapping of sensory inputs to motor outputs, and thus play an important role in decision making. We also show how the absence of expected rewards can switch the attentional bias signal, and thus rapidly and flexibly alter cognitive performance. This theoretical framework incorporates spiking and synaptic dynamics which enable single neuron responses, fMRI activations, psychophysical results, the effects of pharmacological agents, and the effects of damage to parts of the system, to be explicitly simulated and predicted. This computational neuroscience framework provides an approach for integrating different levels of investigation of brain function, and for understanding the relations between them. The models also directly address how bottom-up and top-down processes interact in visual cognition, and show how some apparently serial processes reflect the operation of interacting parallel distributed systems.

Session 3: Biologically Plausible Models for Attention

14:30   Modeling Attention: From Computational Neuroscience to Computer Vision
Fred H. Hamker
Westfälische Wilhelms-Universität, Germany
    Towards a Biologically Plausible Active Visual Search Model
Andrei Zaharescu, Albert L. Rothenstein and John K. Tsotsos
York University, Canada
15:20   coffee break

Session 4: Applications of Attentive Vision

15:50   Visual Attention for Object Recognition in Spatial 3D Data
Simone Frintrop, Andreas Nüchter, and Hartmut Surmann
Fraunhofer AIS Institute, Germany
    AttentiRobot: A Visual Attention-based Landmark Selection Approach for Mobile Robot Navigation
Nabil Ouerhani and Heinz Hügli
University of Neuchatel, Switzerland
    Detection of Frequent Change in Focus of Human Attention from Videos
Nan Hu, Weimin Huang, Surendra Ranganath
Institute for Infocomm Research, Singapore

Poster Session

17:05   On the Usefulness of Attention for Object Recognition
Dirk Walther, Ueli Rutishauser, Christof Koch, and Pietro Perona
California Institute of Technology, CA
    Combining Conspicuity Maps for hROIs Prediction
Claudio M. Privitera, Orazio Gallo, Giorgio Grimoldi, Toyomi Fujita, Lawrence W. Stark
University of California, Berkeley, CA
    A General Purpose Neural Network Simulator for Visual Attention Modeling
Albert L. Rothenstein, Andrei Zaharescu, and John K. Tsotsos
York University, Canada
    Biologically Motivated Selective Attention for Face Localization
Minho Lee and Sang-Woo Ban
Kyungpook National University, South Korea
    Accumulative Computation Method for Motion Features Extraction in Dynamic Selective Visual Attention
Antonio Fernandez-Caballero, María T. López, Miguel A. Fernández, José Mira, Ana E. Delgado and José M. López-Valles
Universidad de Castilla-La Mancha, Spain
    Attentive Object Detection Using an Information Theoretic Saliency Measure
Gerald Fritz, Christin Seifert, Lucas Paletta, and Horst Bischof
JOANNEUM RESEARCH, Austria
18:05   finish