Recent Advances in Computer Vision
Instructor
|
Giovanni Maria Farinella
Università di Catania
Italy
|
|
|
|
Brief Bio
Giovanni Maria Farinella obtained the degree in Computer Science (egregia cum laude) from the University of Catania, Italy, in 2004. He is Founder Member of the IPLAB Research Group at University of Catania since 2005. He was awarded a Doctor of Philosophy (Computer Vision) from the University of Catania in 2008. He is currently a Full Professor at the Department of Mathematics and Computer Science, University of Catania, Italy. His research interests lie in the fields of Computer Vision, Pattern Recognition and Machine Learning, with focus on First Person (Egocentric) Vision. He is Associate Editor of the international journals IEEE Transactions on Pattern Analysis and Machine Intelligence, Pattern Recognition - Elsevier and IET Computer Vision. He has been serving as Area Chair for CVPR 2020/21/22, ICCV 2017/19/21, ECCV 2020, BMVC 2020, WACV 2019, ICPR 2018, and as Program Chair of ECCV 2022, ICIAP 2021 and VISAPP 2019/20/21/22/23. Giovanni Maria Farinella founded (in 2006) and currently directs the International Computer Vision Summer School. He also founded (in 2014) and currently directs the Medical Imaging Summer School. He is member of the European Laboratory for Learning and Intelligent Systems (ELLIS), Senior Member of the IEEE Computer Society, Scientific Advisor of the NVIDIA AI Technology Centre (NVAITC), and board member of the CINI Laboratory of Artificial Intelligence and Intelligent Systems (lead of the area AI for Industry - since 2021). He was awarded the PAMI Mark Everingham Prize 2017. In addition to academic work, Giovanni's industrial experience includes scientific advisorship to different national and international companies and startups, as well as the leadership as Founder and Chief Scientific Officer of Next Vision - Spinoff of the University of Catania.
|
Abstract
In the last decade, consumer imaging devices such as camcorders, digital cameras, smartphones and tablets have been dramatically diffused. The increasing of their computational performances combined with an higher storage capability allowed to design and implement advanced imaging systems that can automatically process visual data with the purpose of understanding the content of the observed scenes. In the next years, we will be conquered by wearable visual devices acquiring, streaming and logging video of our daily life. This new exciting imaging domain, in which the scene is observed from a first person point of view, poses new challenges to the research community, as well as gives the opportunity to build new applications. Many results in image processing and computer vision related to motion analysis, tracking, scene and object recognition and video summarization, have to be re-defined and re-designed by considering the emerging wearable imaging domain. In this tutorial I will give an overview of the recent advances in Computer Vision considering the also wearable domain. Challenges, applications and algorithms will be discussed by considering the state-of-the-art literature.
Keywords
Computer Vision, Wearable Devices, Learning to see, Applications
Aims and Learning Objectives
The objective of this tutorial to provide an overview of the latest advances of Computer Vision also considering challenges and applications in the context of Wearable Imaging Devices. The attendees will become familiar with current and future imaging technologies, as well as with the current state-of-the-art algorithms.
Target Audience
This course is intended for those with a general computing background, and with interest in the topic of image processing, computer vision and machine learning. Ph. D. students, post-docs, young researchers (both academic and industrial), senior researchers (both academic and industrial) or academic/industrial professionals will benefit from the general overview and the introduction of the most recent advances of the field.
Prerequisite Knowledge of Audience
Basic knowledge in the fields Image Processing, Computer Vision, Machine Learning
Detailed Outline
- Introduction and Motivation
- Open Challenges
- State-of-the-Art Algorithms
Image Quality Assessment based on
Machine Learning for the Special Case of Computer-generated Images
Instructor
|
Andre Bigand
ULCO
France
|
|
|
|
Brief Bio
André Bigand received the Ph.D. Degree in 1993 from the University Paris 6 and the HDR degree in 2001 from the Université du Littoral of Calais (ULCO, France). He is currently senior associate professor in ULCO, since 1993. His current research interest include uncertainty modelling and machine learning with applications to image processing and synthesis (particularly noise modelling and filtering). He is currently with the LISIC Laboratory (ULCO). He has 33 years experience teaching and lecturing. He is a visiting professor at UL - Lebanese University, where he teaches "machine learning and pattern recognition" in research master STIP.
|
Abstract
Unbiased global illumination methods based on stochastical techniques provide phororealistic images. They are however prone to noise that can only be reduced by increasing the number of computed samples. The problem of nding the number of samples that are required in order to ensure that most of the observers cannot perceive any noise is still open since the ideal image is unknown. Image quality assessment is well-known considering natural scene images and this is summed up in the tutorial introduction. Image quality (or noise evaluation) of computer-generated images is slightly dierent, since image acquisition is dierent. In this tutorial we address this problem focusing on visual perception of noise. But rather than use known perceptual models we investigate the use of machine learning approaches classically used in the Arti cial Intelligence area as full-reference and reduced-reference metrics. We propose to use such approaches to create a machine learning model based on Learning Machines as SVM, RVM, ... in order to be able to predict which image highlights perceptual noise ([1]). We also investigate the use of soft computing approaches based on fuzzy sets as no-reference metric ([2, 3]). Learning is performed through the use of an example database which is built from experiments of noise perception with human users. These models can then be used in any progressive stochastic global illumination method in order to nd the visual convergence threshold of dierent parts of any image.
This tutorial is structured as a half day presentation (3 hours). The goals of this course are to make students familiar with the underlying techniques that make this possible (machine learning, soft computing).
Trajectories on
Matrix Manifolds for Dynamic 3D Shape Analysis - Applications to Action Recognition and Physical Pain Detection
Instructor
|
Boulbaba Ben Amor
Independent Researcher
France
|
|
|
|
Brief Bio
Boulbaba Ben Amor (IEEE Senior Member) is currently an Associate Professor with the Mines-Telecom Institute and member of the CRIStAL research center (CNRS UMR 9189) in France. He holds an Habilitation (Accreditation to Supervise Research) from the University of Lille1 — Sciences and Technology. He received the Ph.D. and the M.S. degrees in Computer Science, in 2006 and 2003 respectively, both from Ecole Centrale de Lyon (France). In 2002, he graduated with the Engineer degree in Computer Science from the National Engineering School of Sfax (Tunisia). His main interests lie in the area of computer vision and pattern recognition with a focus on human biometrics and behavior understanding. He published several papers in major computer vision journals (IEEE T-PAMI, IEEE T-Cybernetics, IEEE T-IFS, Pattern Recognition, etc.). He served as AC of the WACV 2016 conference, reviewer for many journals (IEEE T-PAMI, IEEE T-IP, IEEE T-IFS, …). He is member of the PC of major conferences (CVPR 2016, ICCV 2015, ICIP and ICPR).
|
Abstract
The analysis of static shapes either 2D and 3D is a well explored research topic with several contributions on non-rigid (elastic) matching of shapes, invariance to rigid transformations, robustness to missing data, recognition, clustering and classification, and so on. How about the analysis of dynamic shapes (i.e. their animations in time)? is there any natural extension? which new chalenges? The aim of this tutoriel is to present existing methodologies and focus on a new trend of modeling dynamic shapes as time-parametrized curves on shape spaces or matrix manifolds. It has several benefits starting from assuming continuity of these curves when developing theories to the implementation of tools useful in processing dynamic shapes. One can cite the definition of a rate-invariant metric to compare shape trajectories, the temporal registration of sequences, filtering and smoothing of trajectories, computing statistical summaries (e.g. sample mean trajectory), alignement of multiple trajectories, etc.). This tutoriel will be oriented to study the topic of human behavior analysis from visual 3D data. With the availability of cost-effective 3D sensors, this topic is nowadays growing with various applications. An illustration of this methodology with two applications (1) action and activity recognition from 3D skeletal shapes and (2) physical pain detection from 4D faces will be in the core of the proposed tutorial