发表日期：2018-08-22 点击击数: 87
报告题目：Facial Analysis: From Biometrics to Healthcare
报告摘要：Facial analysis has witnessed a great progress over the last decades, enabling systems to accurately recognize individuals under controlled environmental conditions. Along the same lines, recent algorithms (e.g. for face) have experienced a high maturity, allowing for the next level: the reliable extraction of semantic attributes. Despite this progress, a number of challenges remain open such as the design of robust algorithms for representing and analysis of features obtained under unconstrained real-world environmental conditions. This talk will present recent algorithmic advancements in semantic attributes, providing information that deviates from the traditional realm of identification, and rather involves other personal and statistical characteristics such as gender, body height, weight, makeup, referred to as soft biometrics. Related benefits include the fast and enrolment free analysis. In addition, the talk will elaborate on recent works concerning face analysis in real-world scenarios and particularly on “Facial behavior analysis in mnemotherapy-sessions for patients with major neurocognitive disorders”. Specifically, approaches will be presented that have been proposed to distinguish between talking, singing, neutral and smiling. Our approaches cater to the challenging setting of current medical recordings of music-therapy sessions, which include continuous pose variations, occlusions, camera-movements, camera-artifacts, as well as changing illumination.
报告人简介：Antitza Dantcheva is a researcher at the STARS team, INRIA, France. Previously, she was Marie Curie and Labex fellow at INRIA and a Postdoctoral Fellow at the Michigan State University and the West Virginia University, USA. She received her PhD in Signal and Image Processing in 2011 from Eurecom / Telecom ParisTech in France. In 2017 she has received the French National Research Agency (ANR) JCJC young researcher grant for her research project "Automated holistic human analysis". She was the recipient of the Best Paper Award (Runner Up) at the IEEE International Conference on Identity, Security and Behavior Analysis (ISBA 2017), the Best Presentation Award at the IEEE International Conference on Multimedia and Expo (ICME 2011), the Best Poster Award at the IAPR International Conference on Biometrics (ICB 2013), as well as the Tabula Rasa Spoofing Award 2013. Dr. Dantcheva is program co-chair of the International Conference of the Biometrics Special Interest Group (BIOSIG) and serves in the editorial board of the journal Multimedia Tools and Applications (MTAP).
Her research interests are in in computer vision, image processing and specifically facial analysis. Her latest results deal with automated healthcare, and specifically with expression recognition for Alzheimer’s disease patients. Other recent results deal with the impact of facial cosmetics on automated gender- and age estimation, as well as face recognition - with focus on impact analysis and on algorithm design, which reduces such impact. Other noticeable results are in the area of soft biometrics for security and commercial applications, where her work was among the first to combine image processing and statistics towards analysis and algorithmic design of soft biometric systems.