KEYNOTE SPEAKERS

 

 

Prof. Jan Sijbers
University of Antwerp, Belgium

Title: Challenges and Solutions in Dynamic X-ray Computed Tomography
Abstract: Most X-ray computed tomography (XCT) experiments assume the object to be stationary. Unvoluntary object motion of imaging system instabilities may however invalidate this assumption, leading to blurred CT images. Alternatively, dynamics may be purposely induced (e.g., compression of the object, foam formation, fluid dynamics, etc). This talk provides a comprehensive overview of those dynamics and suggests possible solutions to compensate for loss of image quality during dynamic CT imaging.

Biodata: Jan Sijbers received a PhD in Physics from the University of Antwerp. In 2010, he was appointed as a senior lecturer at the University of Antwerp. In 2014, he became a full professor. He is Senior Area Editor of IEEE Transactions on Image Processing and Associated Editor of IEEE Transactions on Computational Imaging. Jan Sijbers is the head of imec-Vision Lab and co-founder of IcoMetrix. His main interest are in quantitative image reconstruction, processing, and analysis with focus on Magnetic Resonance Imaging and X-ray Computed Tomography.

Prof. Harald Schenk
Fraunhofer Institute for Photonic Microsystems (IPMS) in Dresden, Germany

Title: Highly Integrated Active Spatial Light Modulators: From Imaging to Holography
Abstract: Several types of analogue deflectable Micro Mirror Arrays (Spatial Light Modulators) were developed and evaluated for image generation in lithography, laser writing and wave front correction. Pitch of the mirrors varies between 16 and 40 µm. The mirrors can be deflected torsional (1D or 2D) or can perform a piston-type deflection when addressed by a voltage. Framerates of 2 kHz and highly repeatable deflection support very fast and precise image generation. Currently, SLMs are being developed for computer generated holography. For this application, high demands need to be met and none of any existing SLM technology meets them fully. Our approach is based on a novel micro mirror architecture allowing to fabricate arrays with 8 million piston-type mirrors of 4 x 6 µm² size.

Biodata: Harald Schenk is a graduate physicist and obtained a doctorate degree from Gerhard Mercator University in Duisburg in Electrical Engineering in 2000. He has been working at Fraunhofer Institute for Photonic Microsystems (IPMS) in Dresden until today. First, he became group leader at Fraunhofer IMS in 2000 and business unit manager in 2002. From 2004 to 2013 he served as Deputy Director at Fraunhofer Institute for Photonic Microsystems (IPMS). In 2013, he was appointed Director of the IPMS and serves as Executive Director since 2021.
He habilitated in Physics at the Brandenburg University of Technology (BTU) in Cottbus in 2008. In 2012, he was appointed full professor for Micro and Nanosystems at the BTU Cottbus where he also headed the Fraunhofer project group MESYS. Since 2019 he is leading the Innovation Campus Electronics and Microsensor Technologies, Cottbus (iCampμs).
Harald Schenk has authored and co-authored more than 35 journal articles, more than 170 conference papers, several book chapters and more than 30 patents or pending patents. He is member of the program and steering committees of several international conferences. For the "Journal for Optical Microsystems " he works as a senior editor. He is a member of SPIE and VDE / VDI.
He is co-founder of the company HiperScan, focusing on near-infrared spectroscopical systems for pharmaceutical applications. Also, he initiated the set-up of and was share-holder of the company Arioso Systems, known for its development of silicon micro loudspeakers.
Main focus of Harald Schenk’s research and development activities is on:
- MEMS / NEMS process development and integration
- Micro and nano actuators and sensors
- Integration of micro sensors and application evaluation

Prof. Raimondo Schettini
University of Milano Bicocca, Italy

Title: The Multiple Facets of Visual Food Recognition in Real-World Scenarios
Abstract: In recent years, several systems have been proposed that utilize advanced machine learning and computer vision techniques to recognize various types of food, to estimate portion sizes, and sometimes even to analyze nutritional content. The outputs of these systems may be used for applications like users’ dietary tracking and monitoring, canteen and restaurant management assistance. Such applications, however, are only possible if these systems can provide reliable information in real scenarios, such as foods from different cuisines, and presentation styles that include complex or mixed dishes.
In this talk, we provide an overview of the different tasks that may exist under the umbrella of "food recognition", highlighting their peculiarities and challenges in real-world scenarios. Finally, we report our experience in building food recognition AI systems for a variety of food-related applications, emphasizing the advantages and disadvantages of the different approaches.

Biodata: Raimondo Schettini is a full professor at the University of Milan-Bicocca, where he has directed the Imaging and Vision Laboratory (www.ivl.disco.unimib.it) since 2003. Previously, he directed the Color Imaging Laboratory of the National Research Council from 1990 to 2002. He has published more than 400 papers and holds 12 patents on signal, image, and video processing, analysis, and classification. As of 2019, he is listed on Stanford University's World Ranking Scientists list for his achievements in artificial intelligence and image processing. He is a member of European Laboratory for Learning and Intelligent Systems (ELLIS), a Fellow of the International Association of Pattern Recognition (IAPR), the Asian-Pacific Artificial Intelligence Association (AAIA), and the International Artificial Intelligence Industry Alliance (AIIA), and member of the advisory board of the international AIQT Foundation. Raimondo Schettini is Chief Technical Officer of the University of Milan Bicocca spin-off 'Imaging and Vision Solutions'. He has chaired numerous international conferences and workshops and is Editor-in-Chief of the international Journal of Imaging.

Prof. Vijayakumar Anand
University of Tartu, Estonia

Title: Recent advances in interferenceless coded aperture correlation holography
Abstract: Interferenceless coded aperture correlation holography (I-COACH) was developed based on the principles of incoherent digital holography and coded aperture imaging in 2017. During the past years, I-COACH has significantly evolved into a valuable imaging technique with surprising capabilities. In this talk, the recent advances in I-COACH and future perspectives will be presented.

Biodata: Prof. Vijayakumar Anand, ERA Chair and Associate Professor of Computational Imaging, Institute of Physics, University of Tartu, Estonia, Adjunct Associate Professor at Swinburne University of Technology, Melbourne, Australia.

Speakers in 2025 to be announced soon......

 

Copyright © 2025 18th International Conference on Machine Vision (www.icmv.org)