The IEEE Biometrics Council invites participants to the upcoming (free)
webinar by Prof. Ran He on “Heterogeneous Face Recognition”. Detail on
the webinar are given below:

Title: Heterogeneous Face Recognition
Speaker: Prof. Dr. Ran He, National Laboratory of Pattern Recognition,
CASIA, China
When: 29 March 2023, at 10am Beijing time (4 am CEST, 9 pm CST, 10pm EST)
Where: Online (Zoom)
Registration: (free, but required):
https://us06web.zoom.us/webinar/register/WN_QhsNRSWRQTOP3KJXkWVVVw

*** Talk Summary ***
Ubiquitous vision sensors not only facilitate the wide application of
face recognition but also generate various heterogeneous sets of facial
images. Matching faces across different sensing modalities raises the
problem of heterogeneous face recognition (HFR) or cross domain face
recognition. Due to significant difference in sensing processes,
heterogeneous images of the same subject have large appearance
variations, which has distinguished HFR from regular visible face
recognition. During last several years, our research group have
investigated a range such problems and developed applications. This talk
will focus on research and recent advances of heterogeneous face
recognition, including fundamental models, face recognition method and
recognition from synthesis.

*** About the Speaker ***
Dr. Ran He received the PhD degrees in pattern recognition and
intelligence system from Institute of Automation, Chinese Academy of
Sciences (CASIA), China, in 2009. He has been a professor at National
Laboratory of Pattern Recognition since December, 2016. He is now
directing the visual perception and machine learning group. He has
published two books and more than 200 papers in refereed journals and
conference proceedings in the areas of computer vision, pattern
recognition, and image processing. He is the editor board member of IEEE
TIP, IEEE T-BIOM and Pattern Recognition. He was the area chair of
CVPR/ECCV/ICML/NeurIPS. His research won IEEE SPS Young Author Best
Paper Award (2020), IAPR ICPR Best Scientific Paper Award (2020), and
IEEE ICB Honorable Mention Paper Award (2019). He is the 2022 recipient
of CAS Outstanding Tutor Award. He is a senior member of the IEEE and
also a Fellow of IAPR.

For more information, visit:
https://ieee-biometrics.org/index.php/activities/webinars

**********************************
The Unconstrained Ear Recognition Challenge 2023 (UERC)

Held in conjunction with IEEE IJCB 2023 https://ijcb2023.ieee-biometrics.org/

Important dates: Registration is open now UERC 2023 Website: http://awe.fri.uni-lj.si/uerc.html
***********************************

*** Motivation ***
Ear recognition is an active area of research within the biometric community. However, the work in this field has long focused on maximizing raw recognition performance, while other aspects critical for the deployment of biometrics recognition techniques in practice have largely been ignored. One such example is demographic bias. Modern ear recognition approaches are not only expected to be highly efficient when recognizing individuals, but also to be equally fair in their decisions, regardless of the demographic characteristics of the subjects, e.g., gender or ethnicity. The 2023 Unconstrained Ear Recognition Challenge will, therefore, investigate the performance as well as demographic bias of existing ear recognition solutions and promote research into bias-mitigation mechanisms that have minimal impact on the recognition performance.

Understanding demographic bias is important because it can help to identify and mitigate inaccuracies and errors in biometric systems, prevent the development of discriminatory systems, and even inform the public policies and regulations related to biometric systems. Research related to demographic bias in ear recognition techniques can help promote the development of more accurate, fair, and just ear recognition systems that are less likely to produce errors or false positives for certain groups of people and protect individual's rights and interests.

To promote research in the bias-aware ear recognition, the Unconstrained Ear Recognition Challenge (UERC) 2023 will bring together researchers working in the field of ear recognition and benchmark existing and new algorithms on a common dataset and under a predefined experimental protocol.

*** Execution ***
UERC 2023 will be organized as a two-track competition, where each track will be focused on one specific goal. Participants will be free to enter only a single track or compete in both. For each track, a dataset, evaluation tool written in Python, and baseline models in Python will be made available. A detailed description of the two tracks is given below:

Track 1: Fair Ear Recognition. The first UERC 2023 evaluation track will collect ear recognition models and score their performance on ear images captured in unconstrained environments. Here, the performance indicators will include both, a measure of recognition performance as well as an estimate of the exhibited demographic bias. Both recognition and bias scores will contribute to the overall ranking. Participants will be free to develop any type of model to maximize performance, while minimizing bias. The final submission for this track will include a working solution (source code or compiled binary), which the organizers will run to evaluate the performance on a sequestered test data.

Track 2: Bias Mitigation. The second UERC 2023 track will address bias mitigation strategies explicitly. Here, a baseline ResNet model (written in Python) will be made available to the participants and the goal will be to design bias mitigation schemes that reduce the initial bias of the models without adversely affecting performance. Such schemes may include additional model blocks and network components, normalization layers, knowledge infusion mechanisms, score normalization procedures, image preprocessing approaches and any other solution capable of reducing bias of the predefined base model. Similarly to the first track, participants will have to submit a working solution that the organizers will evaluate on the sequestered test data.

*** Summary paper and co-authorship ***
The results of UERC 2023 will be published in the IJCB conference paper authored jointly by all participants of the challenge.

*** Organizers ***
+ Asst. Prof. Žiga Emeršič, University of Ljubljana, Faculty of Computer and Information Science, Slovenia
+ Prof. Hazım Kemal Ekenel, Istanbul Technical University, Department of Computer Engineering, Turkey
+ Prof. Guillermo Camara-Chavez, Federal University of Ouro Preto, Brazil
+ Prof. Peter Peer, University of Ljubljana, Faculty of Computer and Information Science, Slovenia
+ Prof. Vitomir Štruc, University of Ljubljana, Faculty of Electrical Engineering, Slovenia, EU

*** Timeline ***
+ February 14th: Promotion of the competition, website draft, registration opens.
+ February 15th: Kick-off of the competition: data, toolkit and instructions made available on UERC  website.
+ April 15th: Possible interim ranking.
+ May 1st: Registration closes, end of the competition.
+ May 15th: Summary paper submission.