|
|
| Special Sessions |
| Special Session 1 |
|
Human Presence Detection for Context-aware Systems |
|
Chair:
Bogdan Raducanu,
Computer Vision Center,
Spain |
|
Scope:
A key advantage of computer vision-based interfaces over the classical ones is their transparency from user's perspective. That's it, vision enable computers to perceive the user, to track his movements and actions. Despite recent developments in computer vision techniques and applications, there is still a long way until our environments will be filled with 'smart devices', that will respond at our demands and intentions. The purpose of the current special session is to offer a meeting opportunity for researchers in different applications areas to interact and exchange ideas so that the concepts presented here can be transfered from one application area to another.
This special session is intended to cover, but it is not limited to, the following topics:
- face detection/tracking/recognition
- facial expression recognition
- human presence detection
- human motion and gesture recognition
- activity interpretation
- multimodal and perceptual user interfaces
- vision-based sensor fusion
- scene modeling and understanding |
|
| |
| Special Session 2 |
|
3D Model Aquisition and Representation |
|
Chair:
Shengyong Chen,
University of Hamburg,
Germany |
| |
|
| Special Session 3 |
|
Beyond Image Enhancement |
|
Chair:
Gabriela Csurka and Marco Bressan,
Xerox Research Centre Europe,
France |
|
Keywords:
Image Aesthetics, Image Quality, Image Enhancement, Image Ranking, Visual Content, Image Data Mining
Scope:
The traditional approach to image enhancement, mostly based on correcting image degradations due to acquisition conditions, sensor noise,
compression, etc., is limited because the final judgment about performance is always subjective. Preference can depend on the semantic content
of the image, contextual features, perceptual priors, aesthetic assumptions such as originality or beauty, emotional response, etc.
The current availability of large amounts of user-labelled data in services such as Flickr provides a unique opportunity for understanding
visual preference more in depth.
This understanding is required to enable applications such as visual content ranking and recommendation systems that will provide access and
value to huge amounts of already available content.
The purpose of the current special session is to offer a meeting opportunity for researchers in different areas to interact and exchange ideas
to better understand these and closely related problems.
Therefore we expect papers related to the following but not exclusive subjects:
- Image aesthetics
- Content-specific image enhancement
- Learning techniques for user preference
- Preference-based image rankink
|
| |
|
| Special Session 4 |
|
Computer Vision Methods in Medicine |
|
Chair:
Joachim Hornegger,
University of Erlangen-Nuremberg,
Germany |
|
|
Scope:
Computer vision is of increasing importance for medical imaging and has significantly changed medical applications and procedures.
C-arm computerized tomography, image guided navigation or augmented reality in medicine are prominent examples that became possible through computer vision technology. The main focus of this special session is on advanced computer vision methods and their application to medicine.
This special session will address, but is not limited to, the following topics:
- Acquisition specific image enhancement
- Geometric calibration of medical devices
- Image guided navigation
- Reconstruction
- Medical image analysis
- Image registration |
| |
|
| Special Session 5 |
|
Bayesian Approach for Inverse Problems in Computer Vision |
|
Chair:
Ali Mohammad-Djafari,
Laboratoire des signaux et systèmes (CNRS-Supélec-UPS),
France |
|
Keywords:
Early Vision, Mono or Multimodal image formation, Image recovery and restoration,
Segmentation and labelling, Tomographic image reconstruction, 3D from 2D, Bayesian approach
Scope:
Many problems of Early Vision, Mono or Multimodal image formation, Image recovery and restoration, Segmentation and labelling, Tomographic
image reconstruction, 3D from 2D, etc. can be presented as inverse problems. ANd, very often, these inverse problems are ill-posed.
The Bayesian approach has shown his need and efficiency for proposing real solutions to these inverse problems.
The main objective of this special session is to regroup new advances and examples of application of this approach to those inverse problems.
|
| |
| Special Session 6 |
|
Mathematical and Linguistic Techniques for Image Mining
|
|
Chair:
Igor Gurevich,
Dorodnicyn Computing Center, Russian Academy of Sciences, Moscow,
the Russian Federation
Heinrich Niemann, University of Erlangen-Nürnberg, Germany
Ovidio Salvetti, Institute of Information Science and Technologies, Italian national Research Council, Italy |
|
Keywords:
Image Mining, Image Analysis, Image Understanding, Image Algebras, Pattern Recognition in Image Understanding, Data Mining in Pattern Recognition, Feature Extraction, Structural and Syntactic Approach, Image Models Acquisition and Validation, Image Knowledge Bases, Image Science Ontologies, Medical Image Analysis
Scope:
Automation of image mining is one of the most important strategic goals in image analysis, recognition and understanding science and technologies. The main subgoals are developing and applying of mathematical theory for constructing image models accepted by efficient pattern recognition algorithms and for standardized representation and selection of image analysis transforms.
The purpose of this special session is to discuss algebraic tools providing a unified language for constructing and implementation of image models and image analysis and recognition transforms. These tools are mainly image algebras, descriptive image algebras and logical and combinatorial pattern recognition algorithms.
Another important task of the session is to discuss linguistic tools for image mining - image knowledge bases and image science ontologies - and to estimate the prospects of the algebraic approach in representation of image analysis knowledge in this environment.
The interpretation of algebraic and linguistic techniques will be illustrated by automated medical diagnosis cases.
This special session is intended to cover, but it is not limited to, the following topics:
- Image Mining
- Mathematical Theory of Image Analysis
- Image Understanding
- Image Algebras
- Pattern Recognition in Image Understanding
- Algebraic, Structural, Syntactic and Logical-Combinatorial Approach
- Image Models
- Image Knowledge Bases
- Image Science Ontologies
- Automation of Medical Image Analysis
|
|
|
|
