Banner
Home      Log In      Contacts      FAQs      INSTICC Portal
 

Keynote Lectures

Self-Supervised Fine-Grained Food Recognition
Petia Radeva, Mathematics and Computer Science, Universitat de Barcelona, Spain

Fingerprint Recognition: Billion-Class Pattern Recognition Problem
Anil Jain, Michigan State University, United States

Causal Discovery from Observation Data: Some Recent Advances
Mario Figueiredo, Instituto de Telecomunicações, Portugal

 

Self-Supervised Fine-Grained Food Recognition

Petia Radeva
Mathematics and Computer Science, Universitat de Barcelona
Spain
 

Brief Bio
Prof. Petia Radeva is a Full professor at the Universitat de Barcelona (UB), Head of the Consolidated Research Group “Artificial Intelligence and Biomedical Applications (AIBA)” at the University of Barcelona. Her main interests are in Machine/Deep learning and Computer Vision and their applications to health. Specific topics of interest: data-centric deep learning, uncertainty modeling, self-supervised learning, continual learning, learning with noisy labeling, multi-modal learning, NeRF, food recognition, food ontology, etc. She is an Associate editor in Chief of Pattern Recognition journal and International Journal of Visual Communication and Image Representation. She is a Research Manager of the State Agency of Research (Agencia Estatal de Investigación, AEI) of the Ministry of Science and Innovation of Spain. She supervised 24 PhD students and published more than 100 SCI journal publications and 250 international chapters and proceedings. Petia Radeva belongs to the top 2% of the World ranking of scientists with the major impact in the field of TIC according to the citations indicators of the popular ranking of Stanford.  Moreover, she was awarded IAPR Fellow since 2015, ICREA Academia’2015 and ICREA Academia’2022 assigned to the 30 best scientists in Catalonia for her scientific merits, received several international and national awards (“Aurora Pons Porrata”, Prize “Antonio Caparrós” ).


Abstract
Food image recognition is a complex computer vision task, because of the large number of fine-grained food classes. In this talk, we will discuss the problems of complex large-scale fine-grained recognition problems as food recognition. We will introduce a novel self-supervised fine-grained multi-task framework and will show its performance on 3 public datasets. We will finish the talk discussing several success stories that open new research lines and professional opportunities for the Computer Vision and Deep Learning community.



 

 

Fingerprint Recognition: Billion-Class Pattern Recognition Problem

Anil Jain
Michigan State University
United States
 

Brief Bio
Anil Jain is a University Distinguished Professor at Michigan State University. His research interests include pattern recognition, computer vision and biometric recognition. He served as the editor-in-chief of the IEEE Transactions on Pattern Analysis and Machine Intelligence and received Fulbright, Guggenheim, Humboldt, and IAPR King Sun Fu awards. He was elected to the United States National Academy of Engineering and The World Academy of Sciences and foreign members of the Indian National Academy of Engineering and the Chinese Academy of Sciences.


Abstract
The skin on human palms and feet has a regular and dense pattern of interlaced ridges and valleys. These ridges, called friction ridges, are claimed to be unique to each finger and do not significantly change over time. It is the uniqueness and persistence properties that have made fingerprint recognition a universally accepted method for person identification by law enforcement agencies for over 100 years. Now, we willingly provide our fingerprints for mobile phone unlock, mobile pay, access control, immigration and national registry. Traditionally, fingerprint representation has been based on minutiae points, points where fingerprint ridges end or bifurcate. Matching two fingerprints involves finding a correspondence between two minutiae sets. While minutiae representation can deliver impressive accuracy, there is a demand to continually improve both the matching accuracy and recognition speed. In this talk, I will present our ongoing work on extracting fixed-length fingerprint representations (embeddings) using deep neural network architectures. Inspired by success of CNN embeddings, we are now extracting minutiae-guided fingerprint embeddings via Vision Transformers (ViT). This fusion of knowledge-driven (minutiae) and data-driven (neural networks) models will likely emerge as the preferred approach to solve complex pattern recognition problems.



 

 

Causal Discovery from Observation Data: Some Recent Advances

Mario Figueiredo
Instituto de Telecomunicações
Portugal
 

Brief Bio
Mário Figueiredo received his PhD (1994) in Electrical and Computer Engineering from Instituto Superior Técnico, University of Lisbon, where he is an IST Distinguished Professor and holder of the Feedzai Chair on Machine Learning. He is a senior researcher and group leader at Instituto de Telecomunicações. His research areas include machine learning, signal processing, and optimization. He received several honors and awards, namely: Fellow of the Institute of Electrical and Electronics Engineers (IEEE), Fellow of the International Association for Pattern Recognition (IAPR), Fellow of the European Association for Signal Processing (EURASIP), W. R. G. Baker Award (IEEE), EURASIP Technical Achievement Award, member of the Portuguese Academy of Engineering, member of the Lisbon Academy of Science. From 2014 to 2018 he was included in the annual list “Highly Cited Researchers”.


Abstract
Causal discovery is an active research field that aims to uncover the underlying causal mechanisms that drive the relationship between a collection of variables and which has applications in many areas, including medicine, biology, economics, and social sciences. In principle, identifying causal relationships requires interventions. However, intervening is often impossible, impractical, or unethical, which has stimulated much research on causal discovery from purely observational data or mixed observational-interventional data. In this talk, after overviewing the causal discovery field, I will discuss some recent advances, namely on causal discovery from data with latent interventions and on what is the quintessential causal discovery problem: distinguishing cause from effect on a pair of dependent variables.



footer