Human Centered Computer Vision

Day - Time: 28 October 2026, h.11:00

Place: Area della Ricerca CNR di Pisa - Room: C-29

Speakers

• Hazim Kemal Ekenel (Department of Computer Engineering at Istanbul Technical University)

Referent

Giuseppe Amato

Abstract

Facial imageanalysis has long been a central problem in computer vision, aiming toautomatically extract rich information about individuals, such as identity,affect, behavior, and other soft biometric traits. Beyond its traditional rolein biometrics, face perception has become a key enabler of human-centered computing, supporting natural and intuitivehuman-computer interaction in applicationsranging from security and smart environments to social robotics and multimediaunderstanding.

In thistalk, I will present an overview of humancentered computer vision research activitiesat theSmartInteraction and Machine Intelligence (SiMiT) Lab of Istanbul Technical University. While maintaining a strong focuson face perception -including facerecognition under challenging conditions, facial expression analysis, andprivacy-preserving techniques such as face anonymization- the talk will also highlight our broader contributionsto human-centered computer vision. These include cross-domain biometrics,robust learning under real-world degradations and occlusions, and generativemodels for image synthesis and manipulation.Furthermore,I will discuss our recent work extending beyond facesand applications in health, assistive technologies,and intelligent environments. Across these directions, a common theme is thedevelopment of robust, adaptable, and trustworthy AI systems that operateeffectively in unconstrained settings and interact seamlessly with humans.