You will find in this announcement the description of this event and how to submit your paper.
the chairs of SS5 / IPTA2020,
Special session title:
Advances in variational image processing
About IPTA 2020 conference:
The tenth International Conference on Image Processing Theory, Tools and Applications IPTA 2020 aims at gathering international researchers, innovators, educators, and practitioners in image processing for attending extensive educational high level materials, sharing their achievements, exchanging their experiences and discussing future orientations. The conference aims to provide an excellent forum to meet and discuss various important issues on image processing and applications. Topics include but are not limited to:
- Image formation, scanning, display and printing
- Image and video processing, coding and compression
- Pattern recognition and computer vision
- Perceptual analysis of images and video
- Image forensics and security
- Machine Learning methods for image and video analysis
- Virtual and augmented reality
- Artificial Intelligence for weather and climate modelling
- Regular paper and special session submission: March 15, 2020 April 5, 2020
- Notification of acceptance: April 15, 2020 May 5, 2020
- Final paper submission and registration: April 30, 2020 May 15, 2020
Variational methods have long been the reference in image processing. These approaches have been widely used to solve ill-posed inverse problems as they propose an explicit modeling framework and may easily embed prior knowledge from experts. With the rise of computing power, deep learning has drastically transformed the research landscape, in particular in image processing and computer vision. However, deep learning approaches still yield mixed results in some applications including (bio)medical imaging as annotated databases are scarcer than in other domains due to the image complexity and the high-level of expertise required.
In this context, variational methods are undergoing a renew as they offer robust and interpretable solutions which are two highly desired properties that are usually missing with deep learning solutions. Especially works introducing hybrid models trying to get the best of both worlds have emerged to give very promising results.
The goal of this special session is to highlight researches in image processing using variational approaches by classic or hybrid strategies. Topics of the session include, without being exhaustive:
- Image processing applications: segmentation, registration, denoising, tracking, deblurring, restoration, etc.
- Classic variational and hybrid (variational/ machine learning) models: model-based, plug-and-play, deep unfolding, etc.
- All application fields are welcome (medical/biomedical, remote sensing, computer vision, etc.).
The best paper of this special session may be invited to publish an extended version to one of the following journals:
- Signal Processing: Image Communication, Elsevier ;
- Innovation and Research in BioMedical Engeneering, Elsevier ;
- International Journal of Signal and Imaging Systems Engineering, Inderscience.
Special session chairs:
- Noémie Debroux, Institut Pascal (UMR6602), Le Puy-en-Velay, France
- Marie-Ange Lebre, Institut Pascal (UMR6602), Le Puy-en-Velay, France
- Odyssée Merveille, CREATIS (UMR5220), Lyon, France
- Antoine Vacavant, Institut Pascal (UMR6602), Le Puy-en-Velay, France
Papers must be submitted in English and formatted according to the instructions given in the website above. Please check and carefully follow the instructions provided in the templates which contains detailed instructions on formatting your document.
Authors should submit a paper maximum of 6 pages including figures and references, carefully checked for correct grammar and spelling, using the on-line submission procedure indicated below. Authors are required to include their names and affiliations in their papers.
All accepted papers will be submitted for inclusion in IEEE Xplore and will be submitted for indexing by EI.
A selection of the best papers from the conference will be invited to submit an extended version to special issues of journals.