Our paper titled "Averager Student: Distillation from Undistillable Teacher" is accepted by Tiny Papers @ ICLR 2023. In this paper, with the aim of contributing to solutions for model intellectual property (IP) protection, a novel approach is proposed, which improves the distillation from an undistillable teacher whose goal is make the distillation difficult for students, with the purpose of model protection.
We are pleased to announce the 8th Science Academy Machine Learning Summer School, which will be held in Suleyman Demirel Cultural Center in Istanbul Technical University between 22-25 June this year. Istanbul Technical University Informatics Institute organizes this summer school under the supervision of the Science Academy. Read more here.
The International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM) is the annual workshop organized by the working group Multimedia Understanding through Semantics, Computation and Learning (MUSCLE) of the European Research Consortium for Informatics and Mathematics (ERCIM). This year, IWCIM takes place as a satellite workshop to IEEE ICASSP 2023, to be held in Rhodes Island, Greece on June 4 - 9, 2023. Read more here.
Reyhan Kevser Keser is the leading author of a recent paper on knowledge distillation titled as “PURSUhInT: In Search of Informative Hint Points Based on Layer Clustering for Knowledge Distillation” which is available online in 'Expert Systems with Applications' journal. Read more here.
Onur Can Koyun is the leading author of a recent paper on small object detection titled as “Focus-and-Detect: A small object detection framework for aerial images” which has been published in 'Signal Processing: Image Communication' journal. Read more here.
The M.S. thesis, “Rotational Duplicate Layer: A New and Compact Convolutional Neural Network Layer” by Mr. Erdem Safa Akkul was chosen to be the best thesis in terms of thesis outcomes among all ASELSAN Akademi theses across all disciplines in 2022. Outcomes of Mr. Akkul’s thesis work include two international conference papers and one national patent disclosure. Mr. Akkul’s thesis was supervised by Prof. Behçet Uğur Töreyin.
Research assistants Reyhan Kevser Keser and Onur Can Koyun represented İTÜ at SAHA EXPO Fair 2022. They represented İTÜ with the project titled “Developing Video Object Detection Methods Based on Deep Learning” which is supervised by Prof. Behçet Uğur Töreyin and funded by ASELSAN.
The invention of the team, including group members Sibel Çimen Yetiş, Dursun Ali Ekinci, Abdulkerim Çapar and Behçet Uğur Töreyin, numbered TR 2019/19929, titled "Method and System for Automated Myelin Detection in Fluorescent Microscopy Images Using Machine Learning" was granted by the Turkish Patent and Trademark Office.
Over the recent years, IoT is gaining importance for its capacity of building a hyperconnected world with minimized human interaction. Seamlessly, the giant connection of things provides an improved platform for the integration of different devices to address explicit needs. Since this technology is upgrading in a rapid phase, it paves a way for an autonomous IoT infrastructure with ambient intelligence. The combination of IoT and computer vision technology could support many applications like smart transportation, intelligent healthcare monitoring, online education, smart manufacturing, etc. Specifically, after this technological breakthrough intelligent transportation has obtained maximized benefits from them. Researchers, scholars, academicians, and other industry stakeholders from this area are welcome to submit their high-quality research works for improving this current transportation infrastructure. Thus, the purpose of this special issue is to invite both academic and industrial participants to deliver innovative solutions for the betterment of this domain. Read more here.
This special session focuses on microscopy image analysis. Researchers are encouraged to submit their original contributions on image processing and computer vision based 2D, 3D, 2D+t, 3D+t (semi-)automated data interpretation techniques for images acquired by different types of microscopes. This year, TIPTEKNO 2022 takes place in Bebek, Turkey – 31 October - 2 November, 2022. Read more here.
This workshop aims to shorten the gap between edge-based and cloud-based approaches for multimedia understanding by addressing the challenges and opportunities of the combined approach. Relevance will be given to transversal issues, such as those related to the privacy and security of such hybrid strategies, as well as to the quest for optimal and greener configuration of AI services towards sustainable and ethical use of computing resources. This year, SITIS 2022 takes place in Dijon, France – October 19-21, 2022. Read more here.
The International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM) is the annual workshop organized by the working group Multimedia Understanding through Semantics, Computation, and Learning (MUSCLE) of the European Research Consortium for Informatics and Mathematics (ERCIM). This year, IWCIM takes place as a satellite workshop to ICCCI 2022, to be held in Hammamet, Tunisia on September 28 - 30, 2022. Read more here.
Dr. Abdulkerim Çapar, one of the members of our research group, made a presentation on "Artificial Intelligence in Genetic Diagnosis" at the "Artificial Intelligence in Health" symposium organized by the Turkish Health Institutes (TÜSEB). Click here for more information.
Artificial intelligence-based genetic diagnosis systems developed by Argenit, of which Dr. Abdulkerim Çapar - member of our research group - is a partner, were reported in the Sabah newspaper.
Due to recent advances in artificial intelligence, image and video analysis, and machine learning techniques, as well as progress in enabling technologies, such as semiconductor device design and fabrication, volumetric fire detection methods yield superior results, as compared to other methods and point detectors. These developments pave the way for embedded computer vision systems equipped with lenses and cameras attached to computing platform which process an acquired sequence of images. The Journal of Imaging Editorial Board launches this call for papers for a Special Issue on “Applications of Artificial Intelligence and Computer Vision on Fire and Smoke Detection”. Deadline for manuscript submissions: 15 October 2021. Read more here.
The International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM) is the annual workshop organized by the working group Multimedia Understanding through Semantics, Computation, and Learning (MUSCLE) of the European Research Consortium for Informatics and Mathematics (ERCIM). This year, IWCIM takes place as a satellite workshop to ICCCI 2021, to be held in Rhodes, Greece on September 28 - October 1, 2021. Read more here.
The March session of the ATHENS (Advanced Technology Higher Education Network/ Socrates) program was carried out online by the member institutions from 15-19 March 2021. Istanbul Technical University has been a member of the ATHENS network since 2007 (http://athensnetwork.eu/). Dr. Töreyin offered the course "ITU INF21 - Introduction to Artificial Intelligence and Machine Learning” in the ATHENS week. As part of the course, two guest lecturers, Dr. Carly Eckert and Dr. Muhammad Aurangzeb Ahmad delivered a talk on “Fairness in Healthcare Al: Challenges and Opportunities”. Dr. Carly Eckertis a preventive medicine physician and epidemiologist and she has worked in the healthcare AI industry for 4 years, collaborating closely with data scientists developing and implementing machine learning solutions in healthcare. Dr. Muhammad Aurangzeb Ahmad is an Affiliate Assistant Professor in the Department of Computer Science at the University of Washington and a Research Scientist at KenSci, an AI in healthcare company based in Seattle, USA. The recording of the lecture can be accessed here. Slides of the lecture can be downloaded here.
Beltus Nkwair is selected as the winner for the "COMSTECH-ICYF Essay Competition on Artificial Intelligence" The Organization of Islamic Cooperation Standing Committee on Scientific and Technological Cooperation (COMSTECH) and the Islamic Cooperation Youth Forum (ICYF) organized an international essay competition on Artificial Intelligence (AI) open to students and professionals from all OIC countries (57 member states). [Link: http://www.comstech.org/icyf/ ]. Representing Istanbul Technical University under the advanced students and professionals category (MSc and Ph.D.) was Beltus Nkwawir Wiysobunri, a graduate student, from Cameroon, in Computer Science Program at Informatics Institute, and a member of our research group. His essay titled "A Deep Learning Automatic Tactile Pavement Detection System for the Visually-Impaired” was selected as one of the top three winning essays in the Professional Category of the COMSTECH-ICYF Essay Competition on Artificial Intelligence. The Award ceremony is scheduled to be held online in mid-February 2021. The award-winning essay presenting the project idea by Beltus can be accessed here.
Dr. Töreyin delivered a talk (in Turkish) about machine learning and artificial intelligence at the Personal Data Protection Authority of Turkey (Kişisel Verileri Koruma Kurumu), on "Data Privacy Day", Jan. 28th, 2021. The talk can be accessed here. Slides of the talk are also avaiable.
For many of us, the year 2020 was arguably one of the hardest ones in our lives. Still, it was a productive year for the Signal Processing for Computational Intelligence Research Group members. Our group members defended one Ph.D. thesis and six masters of science theses in the past year. We have one patent application granted, twelve papers published and/or accepted for publication in peer-reviewed journals and conference proceedings. Three group members received national awards and international recognition. Our research continues to receive attention all around the world. Detailed list of research outcomes and recognitions can be accessed here. At SP4CING, we strive to produce more and better research outcomes in terms of quality and quantity in 2021. We are indebted to Istanbul Technical University and all organizations that fund our research. Wish you all a healthy and happy new year.