Biometrics and Machine Learning Group
Latest news
We are pleased to announce that Weronika Gutfeter defended her doctoral dissertation on face recognition based on multi-shot images using deep aggregation networks, supervised by prof. Andrzej Pacut, on the 18th of May, 2023.
We are pleased to announce that on November 17, 2022, Ewelina Bartuzi-Trokielewicz successfully defended her doctoral dissertation titled „Presentation attack-resistant palm recognition for mobile devices in unconstrained conditions”. Her research was conducted under the supervision of Prof. Andrzej Pacut.
We are pleased to announce that Mateusz Trokielewicz defended (with honors) his doctoral dissertation entitled „Iris Recognition Methods Resistant to Biological Changes in the Eye” , supervised by prof. Czajka and prof. Pacut, on the 18th of July, 2019.
Iris scanner can distinguish dead eyeballs from living ones: MIT Technology Review reports on our recent developements in the field of presentation attack detection for cadaver irises.
BIOWIZ
Design and construction of a system for recognition of persons (offenders) based on face images captured on photograph or video material
Financing
NCBR (National Center for Research and Development) - Research and development projects for state defence and security 7/2015, contract no. DOB-BIO7/18/02/2015
The consortium
• Warsaw University of Technology, Institute of Control and Computation Engineering
• AGH University of Science and Technology in Cracow
• Research and Academic Computer Network - research institute
• Polish Platform for Homeland Security
Amount granted
Warsaw University of Technology: 1 355 815 PLN
Timeframe
20-12-2015 – 20-10-2017
Project description
The goal of this project is to build a system for biometric identification of perpetrators
of offences or criminals based on photographs and/or video materials. The biometric part
of the system will consist of integrated modules, including face detection module,
surveillance module, “biometric engines” for face and silhouette recognition, and fusion
module generating biometric profiles. Biometric modules will be integrated with a
database, which will integrate the biometric data with the police records. The system is
thought as an interactive tool and will be operating in various application scenarios,
including face detection, isolation of video frames containing faces, surveillance in video
materials and identification of persons marked on photo and video materials using the
biometric profiles. Modular construction enables for easy supplementing the scenario list
and actualization of biometric techniques. The system will be an indispensable tool for
personal identification tasks.
Keywords
biometrics, identity identification, face detection, tracking, silhouette recognition