LECCE – Seventeen workshops, eleven seminars, four keynote speeches, presentation of approximately 150 works and other special events with more than 400 professors, researchers and researchers from or affiliated around the world: Monday 23 to Friday 27 May the twenty-first edition of ICIAP – International conference on image analysis and processing.
The Palmieri Boarding School and the Apollo Theater in the heart of Salento, the capital, will host a rich program to explore the latest trends in computer vision and artificial intelligence. The biennial travel conference (postponed due to the pandemic in 2021) is hosted in Lecce by the Isasi – CNR Institute for Applied Sciences and Intelligent Systems under the general coordination of Stan Sclaroff (Department of Electrical and Computer Engineering) at Boston University Cos and Engineering) (Institute of Applied Sciences and Intelligent Systems – Cnr) and is organized by Cvpl – Italian Association for Computer Vision Research, Pattern Recognition and Machine Learning and by Iapr – International Association for Pattern Recognition.
ICAP – which for the first time can also be watched entirely online – therefore focuses mainly on trends in computer vision and image processing, ranging from the most theoretical to the practical aspects. Artificial intelligence applied to image analysis and recognition will be discussed in a wide range of presentations on the latest research results, the deepening of security applications (drones, smart cities, production processes) and then the exploration of multidimensional systems, 3D computer vision, digitization, security and preservation of cultural heritage. We will also talk about deepfake contrast, face-filming, medical imaging related to the fight against covid-19, through CT scan and chest x-rays, tumor prevention and monitoring of premature babies. touching areas that have not yet been explored. Presentations will be attended by Italian speakers from the United Arab Emirates, Austria, Belgium, Brazil, Canada, Switzerland, China, Czech Republic, Germany, Spain, Finland, France, Great Britain, Georgia, Greece, Japan, South, online. , Sri Lanka, Luxembourg, Latvia, The Netherlands, New Zealand, Peru, Poland, Romania, Russia, Sweden, Turkey, Taiwan, United States and South Africa.
The program will start on Monday 23 and Tuesday 24 May (from 9 to 18) at Convitto Palmieri with two full days of workshops and tutorials that will give participants the opportunity to hear and discuss a variety of topics. From Wednesday 25 to Friday 27 (from 9 to 18:30) the conference will be broadcast live at the Apollo Theater with the rapid presentation of works presented and selected from around the world through short speeches (fifteen minutes) or “virtual” posters . (in which you compose your ideas). Three “competitions” dedicated to deeply false images, ruminant parasites and the use of robots in the field of Agriculture.
Wednesday 25 at 8:30, after greetings by Cosimo Distante (Cnr-Isasi) and Stan Sclaroff (Boston University), the opening will be given to Roberto Cipolla, a British computer vision researcher and professor of Computer Engineering at Cambridge University. , who will talk about “Geometry, uncertainty and deep learning”. The teacher will provide a brief overview of some of the basic principles with examples of reconstruction, registration and recognition of 3D objects and their translation into new commercial applications. In the afternoon instead of Larry S. Davis, a professor at the University of Maryland, director of the Center for Automated Research (CfAR), an Amazon consultant who will talk about computer vision applications in the fashion world.
In fact, “virtual test” solutions are being developed. Taking a picture of the person, you can observe how the selected clothes are dressed thanks to complex three-dimensional body estimation algorithms.
Thursday 26, the day will always open at 8:30 a.m. with the intervention of Spanish researcher Laura Leal-Taixe, a professor at the University of Munich, who in 2017 received the Sofja Kovalevskaja Prize, a scientific recognition awarded for the work of SocialMaps. Alexander von Humboldt Foundation. His research interests are the dynamic understanding of the scene, especially the monitoring and segmentation of multiple objects, as well as the machine learning for video analysis.
Friday 27 at 8:30 the last Keynote will be assigned to Dima Damen, a professor at the University of Bristol, who will talk about “video comprehension” but from an “egocentric point of view”, ie the analysis of first-person films. The teacher’s research interests are, in fact, in the automatic understanding of the interactions, energies and activities of objects using static and portable optical sensors (and depth).
Thursday 26 from 16:30 to 18:30 also appointment with the national assembly of CVPL – Italian Association for Computer Vision Research, Pattern Recognition and Machine Learning with the participation of President Gian Luca Foresti, full professor in the department of Mathematics, Computer Science and Physics (Dmif) of the University of Udine.
Isasi (formerly known as the Institute of Government) was founded in 1968 by the pioneer of artificial intelligence in Italy Edoardo Renato Caianello and over the years has produced numerous scientific innovations in the field of artificial intelligence and automated systems. In particular, the Lecce unit specializes in artificial intelligence, image processing, automatic recognition systems and the ethical aspects associated with them. Cnr is a public body and aims to conduct, promote, disseminate, transfer and improve research activities in key areas of knowledge development and its applications for the scientific, technological, economic and social development of the country.
The twenty-first edition of ICAP – International Conference on Image Analysis and Processing is promoted in collaboration with the Region of Puglia, the province of Lecce, the Museum Center of Puglia, the Library of the Museum of Lecce, the Springer Lncs with the support of Nvidia, E4 computer engineering, ImageS.