For the sixth International Conference on Image Processing Tools, Theory and Applications (IPTA) being held in Oulu, Finland, between December 12 and 15, 2016, we are looking for researchers desiring to organize Special Sessions on critical and emerging topics in image processing. Each Special Session is expected to count between 4 and 8 invited speakers. Presentations will be oral.
In advance of the meeting, invited contributors are required to submit their papers for inclusion in the proceedings (which will undergo normal peer review).
If you are interested, please submit a proposal to Prof. William Puech (William.Puech "AT" lirmm.fr) and Prof. Eric Granger ( Eric.Granger "AT" etsmtl.ca ) , and cc to Adj. Prof. Abdenour Hadid (hadid "AT" ee.oulu.fi) using the following template:Key Words:
Special Session Proposals should be submitted as soon as possible, but no later than May, 31st, 2016.
Special sessions participants are encouraged to submit their papers for inclusion in the proceedings. They will get full review (with the possibility of being rejected).
Accepted Special Sessions:
1. Multiple Instance Learning for Image Analysis
Special Session Chair: Marc-André Carbonneau
Submission Deadline : June 15th, 2016 July 15th, 2016 (Extended)
Aims & topics:
Multiple instance learning (MIL) is a form of weakly-supervised learning. In this setting, instances are organized in bags, and a label is provided for the bags but not the individual instances. Many problems can be formulated a MIL, especially in image and video recognition. Typically, images are divided in parts, the instances, which will be grouped in bags representing the whole image. With MIL, it is possible to learn from images loosely annotated. This is particularly useful in contexts where manual segmentation and local annotation is costly (e.g. medical imaging). MIL can also be used in contexts where an object or a place has to be classified from a collection of pictures (e.g. travel sites). In content video analysis, it enables the recognition of action from scene transcripts. The proposed special session presents recent MIL algorithms for image analysis.
- Multiple instance learning
- Set classification
- Anomaly detection
- Weakly supervised learning
- Image classification
- Bag-of-words methods
- Object localization and tracking in images
- Group-based classification
For further information, please contact: marcandre"DOT"carbonneau"AT"gmail"DOT"com
2. Deep learning for image analysis and classification
Special Session Chair: Zhaoqiang Xia
Submission Deadline : June 15th, 2016 July 15th, 2016 (Extended)
Aims & topics:
Deep learning is an emergent field of machine learning focusing on learning representations of data. The deep learning techniques learn a deep architecture to represent the data and achieve promising performances which the existing shallow models cannot obtain. Deep learning has recently found success in the field of image analysis and classification. Deep Learning’s power comes from learning rich representations of data that can be tuned for the task of interest, such as the image dehazing, face recognition and so on. Traditional features used in the machine learning techniques can be easily replaced by the deep learning models. In this special session, we encourage submissions that effectively deploy Deep Learning to advance the state of the art across the domain of image analysis and classification.
- Deep learning
- Image analysis
- Image classification
- Image quality assessment
- Image dehazing
- Object localization and tracking in images
- Face and emotion recognition
For further information, please contact: zxia"AT"nwpu"DOT"edu"DOT" cn
3. Biometrics and Forensic Investigations
Co-organized by EURASIP SAT on "Biometrics, Data Forensics, and Security"
Special Session Chairs: Paulo Lobato Correia, Abdenour Hadid
Submission Deadline : June 15th, 2016 July 15th, 2016 (Extended)
Aims & topics:
The special session focuses on biometric recognition techniques and their application in the context of forensic investigations. The question of identifying or verifying the identity of people allegedly involved in some action is increasingly relevant. In this context forensics and biometrics techniques are often involved and a relationship between them exists. The cooperation between these two research areas can facilitate the identification of people involved in criminal actions or civil incidents. This session addresses the joint efforts between these two communities, which have traditionally operated in relative isolation from each other. Important synergies can result from bridging the gap between biometrics and forensics, leading to the development of novel solutions to important forensic problems.
- Audiovisual biometrics for multimedia forensics
- Forensic examination and soft biometrics
- Forensic behavioural biometrics
- Biometric analysis of crime scene traces and their forensic interpretation
- Combination of multimodal biometrics with other forensic evidence
- Biometric-based cybercrime investigation
- Forensic investigation using big-data
- Biometric data de-identification
- Ethical and societal implications of emerging forensics biometrics
- Paulo"DOT" Correia"AT" lx"DOT" it"DOT" pt,
- hadid"AT" ee"DOT" oulu"DOT" fi
4. Anisotropic Filters, locally-adaptive diffusion and Scale-Space
Special Session Chair: Philippe Montesinos
Submission Deadline : June 15th, 2016 July 15th, 2016 (Extended)
Aims & topics:
This track focuses on Anisotropic Filtering in the context of Scale Space. Several domains of image processing are concerned with image filtering: image segmentation of edges, crest-lines, corners, textures, also image restoration and noise removal, etc.
Since 90s, with the development of the scale space theory, several other important spaces have been defined. Most of the PDE (Partial Derivative Equation) numerical schemes describing these spaces are driven by the direction of the gradient which is obtained generally by isotropic filters or tensor anisotropic filtering. Recently, more and more research works address the problem of the directional description of the image surface, thus obtaining gradient and its orientation using anisotropic filters. These new techniques enables a more complete image surface description than these obtained with isotropic scale space. They can define several different characteristic orientations at each pixel for example for texture segmentation. Moreover these techniques, strongly linked to scale-spaces, could lead to image description combining orientations and scales. In the field of image enhancement, image surface description at multiple orientations enables the definition of new restoration schemes. This special session focuses on these new descriptions and on methods using such descriptions.
- Anisotropic filtering
- Scale-space
- PDE
- Tensor filtering
- Multiple orientations
- Image segmentation using anisotropic filters
- Causal filters
- Hybrid Filters
5. Advances in Medical Imaging
Special Session Chairs: Su Ruan, Rachid Jennane, Mohammed El Hassouni, Odemir Bruno
Submission Deadline : June 15th, 2016 July 15th, 2016 (Extended)
Aims & topics:
Imaging has become an essential component in many fields of medical and biological image analysis, and clinical practice. Image processing tools enable quantitative analysis and visualization of medical images of numerous modalities such as PET, MRI, CT, X-ray, microscopy, ultrasound, hyperspectral imaging. The recent advances in medical imaging have revolutionized the diagnostic accuracy of the medical images.
As image processing techniques are at the core of the diagnosis, competent and reliable algorithms that can either reduce the procedure time and/or aid in the diagnosis/detection of disease are highly encouraged and requested.
This Special Session aims to present advances and original methods of image processing and applications in the field of health.
The session welcomes papers on the following research topics (but not limited to):
- Image restauration
- Image segmentation and classification
- Image registration
- Feature extraction
- Texture analysis
- Hyperspectral analysis
- Machine learning methods in medical imaging
- Computer Aided Diagnosis (CAD) systems
- Big data in medical imaging
- Visualization techniques
- Shape and motion measurements
- Inter-study and inter-subject registration
- Longitudinal / temporal studies
- Su Ruan (su"DOT"ruan"AT"univ-rouen"DOT"fr)
- Rachid Jennane (Rachid"DOT"Jennane"AT"univ-orleans"DOT"fr)
- Mohammed El Hassouni (mohamed"DOT"elhassouni"AT"gmail"DOT"com)
- Odemir Bruno (bruno"AT"ifsc"DOT"usp"DOT"br)