(last update of this page: May 2024)
Resources (Research)
Find our recent releases of code or data at https://github.com/HalmstadUniversityBiometrics
Published Software
EFaR-Efficient-Face-Recognition (2023)
Software code for pose-invariant face recognition based on lightweight deep convolutional models of just a few Mbs, suitable for mobile recognition
Models submitted to EFaR 2023: Efficient Face Recognition Competition (see competitions section of this web)
Joint publication by organizers and participants describing the competition: J. N. Kolf, F. Boutros, J. Elliesen, M. Theuerkauf, N. Damer, M. Alansarir, O. A. Hay, S. Alansari, S. Javed, N. Werghi, K. Grm, V. Struc, F. Alonso-Fernandez, K. Hernandez Diaz, J. Bigun, A. George, C. Ecabert, H. O. Shahreza, K. Kotwal, S. Marcel, I. Medvedev, J. Bo, D. Nunes, A. Hassanpour, P. Khatiwada, A. A. Toor, B. Yang, “EFaR 2023: Efficient Face Recognition Competition”, Proc. IEEE/IAPR International Joint Conference on Biometrics, IJCB, Ljubljana, Slovenia, Sep 25-28, 2023 https://arxiv.org/abs/2308.04168
Paper describing the basis of training the submitted methods: F. Alonso-Fernandez, J. Barrachina, K. Hernandez Diaz, J. Bigun, “SqueezeFacePoseNet: Lightweight Face Verification Across Different Poses for Mobile Platforms”, Proc. IAPR TC4 Workshop on Mobile and Wearable Biometrics, WMWB, in conjunction with IEEE/IAPR Intl Conf on Pattern Recognition, ICPR, 2020, Milan, Italy, 10-15 January 2021 (online) (Oral) https://doi.org/10.1007/978-3-030-68793-9_10, https://arxiv.org/abs/2007.08566, https://youtu.be/suJmO8IWp8k (collaboration due to visit to the company FacePhi Biometria in Spain)
Available: https://github.com/HalmstadUniversityBiometrics/EFaR-Efficient-Face-Recognition
FaceDancer: Pose- and Occlusion-Aware High-Fidelity Face Swapping (2023)
Software code for subject-agnostic face swapping and identity transfer
Model submitted to DFGC 2022: The Second DeepFake Game Competition (see competitions section of this web)
Based on work of co-supervised PhD student Felix Rosberg
Described in paper: F. Rosberg, E. Aksoy, F. Alonso-Fernandez, C. Englund, “FaceDancer: Pose- and Occlusion-Aware High Fidelity Face Swapping”, Proc. IEEE/CVF Winter Conference on Applications of Computer Vision, WACV, Waikoloa, Hawaii, Jan 3-7, 2023 http://arxiv.org/abs/2210.10473
Available: https://github.com/HalmstadUniversityBiometrics/FaceDancer
Lightweight Face Recognition Models (2020)
Software code for pose-invariant face recognition based on lightweight deep convolutional models of just a few Mbs, suitable for mobile recognition
Described in paper: F. Alonso-Fernandez, J. Barrachina, K. Hernandez Diaz, J. Bigun, “SqueezeFacePoseNet: Lightweight Face Verification Across Different Poses for Mobile Platforms”, Proc. IAPR TC4 Workshop on Mobile and Wearable Biometrics, WMWB, in conjunction with IEEE/IAPR Intl Conf on Pattern Recognition, ICPR, 2020, Milan, Italy, 10-15 January 2021
Available at https://github.com/HalmstadUniversityBiometrics/SqueezeFacePoseNet-Lightweight-Face-Recognition
Iris Segmentation Code (2015)
Software code for iris segmentation based on the Generalized Structure Tensor (GST)
Described in paper: F. Alonso-Fernandez, J. Bigun, “Iris Boundaries Segmentation Using the Generalized Structure Tensor”, Proc. Intl Conf on Biometrics: Theory, Apps and Systems, BTAS, Washington DC, September 23-26, 2012
Available at https://github.com/HalmstadUniversityBiometrics/Iris-Segmentation-GST-Matlab
Published Databases
Participation in the design, collection, supervision, management, integration and distribution of several databases. In several cases, the work is the result of collaborative European or national (Spanish) public projects. Other cases include the engagement of MSc students in the acquisition, as part of their research. All databases are available (or will be made available) to the research community. In most cases, there is a publication describing the database, which is usually co-written by all participating institutions.
LFW-Beautified: A Dataset of Face Images with Beautification and Augmented Reality Filters (2022)
Dataset containing enhancement filters (which mostly modify contrast and lightning) as well as augmented reality filters that incorporate items like “Dog nose”, “Transparent glasses”, “Sunglasses with slight transparency”, and “Sunglasses with no transparency” to the face image. Additionally, images with sunglasses are processed with a reconstruction network which has been trained to learn to reverse such modifications. A total of 34592 annotated images.
Generated in the context of the M.S. thesis of Pontus Hedman, Vasilios Skepetzis, “The Effect of Beautification Filters on Image Recognition: Are filtered social media images viable Open Source Intelligence?” (http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-44799)
Experiments with the database described in the paper: P. Hedman, V. Skepetzis, K. Hernandez-Diaz, J. Bigun, F. Alonso-Fernandez, “On the Effect of Selfie Beautification Filters on Face Detection and Recognition”, Pattern Recognition Letters, Elsevier, Special Issue on Mobile and Wearable Biometrics, https://doi.org/10.1016/j.patrec.2022.09.018, https://arxiv.org/abs/2110.08934
Database described in the paper: P. Hedman, V. Skepetzis, K. Hernandez-Diaz, J. Bigun, F. Alonso-Fernandez, “LFW-Beautified: A Dataset of Face Images with Beautification and Augmented Reality Filters”, https://arxiv.org/abs/2203.06082
Research presented in two invited talks at the Workshop of the Biometrics Working Group of TeleTrusT, the IT Security Association in Germany, Darmstadt, Germany, September 12, 2022 (https://eab.org/events/program/273) and the Workshop of the Norwegian Biometric Forum (NBF), Oslo, Norway, October 13, 2022 (https://eab.org/events/program/298), both in coop. with the European Association for Biometrics (EAB)
Available at https://github.com/HalmstadUniversityBiometrics/LFW-Beautified
Database of images for drone detection (2020)
Dataset containing IR, visible and audio data that can be used to train and evaluate drone detection sensors and systems. The dataset contains 90 audio clips and 650 videos (365 IR and 285 visible). If all images are extracted from all the videos the dataset has a total of 203328 annotated images. Video labels: Airplane, Bird, Drone and Helicopter. Audio labels: Drone, Helicopter and Background.
Generated in the context of the MSc thesis of Fredrik Svanström, “Drone detection and classification using sensor fusion and machine learning” (http://urn.kb.se/resolve?urn=urn:nbn:se:hh:diva-42141)
Experiments with the database reported in the papers:
F. Svanström, C. Englund, F Alonso-Fernandez, “Real-Time Drone Detection and Tracking With Visible, Thermal and Acoustic Sensors”, Proc. IEEE/IAPR Intl Conf on Pattern Recognition, ICPR, Milan, Italy, 10-15 January 2021 (online) https://arxiv.org/abs/2007.07396
F. Svanström, F Alonso-Fernandez, C. Englund, “Drone Detection and Tracking in Real-Time by Fusion of Different Sensing Modalities”, Drones, 2022, 6(11), 317 http://arxiv.org/abs/2207.01927, https://doi.org/10.3390/drones6110317
Database described in detail in the paper: F. Svanström, F Alonso-Fernandez, C. Englund, “A Dataset for Multi-Sensor Drone Detection”, Data in Brief, Elsevier, vol. 39, Dec 2021 https://arxiv.org/abs/2111.01888. https://doi.org/10.1016/j.dib.2021.107521
Available at https://github.com/DroneDetectionThesis/Drone-detection-dataset
Database of ground-truth iris segmentation (2014)
Segmentation data of nearly 10000 iris images, in conjunction with Salzburg University (Austria), with the objective of enabling accurate evaluation of iris segmentation algorithms.
Described in the paper: H. Hofbauer, F. Alonso-Fernandez, P. Wild, J. Bigun, A. Uhl, “A Ground Truth for Iris Segmentation”, Proc. IEEE/IAPR 22nd International Conference on Pattern Recognition, ICPR, Stockholm, August 24-28, 2014
Available at https://github.com/HalmstadUniversityBiometrics/Iris-Segmentation-Groundtruth-Database
ATVS-FakeIris Database (ATVS-FIr DB)
Development and acquisition with two supervised MSc students (800 real iris and associated fake image samples). Real samples are from the BioSec database (below), captured with a LG Iris Access EOU3000. Fake samples acquired with the same sensor from high quality printed images of the original samples.
Described in the paper: V. Ruiz-Albacete, P. Tome-Gonzalez, F. Alonso-Fernandez, J. Galbally, J. Fierrez, and J. Ortega-Garcia, “Direct attacks using fake images in iris verification”, Proc. COST 2101 Workshop on Biometrics and Identity Management, BIOID, Roskilde University, Denmark, 7-9 May 2008
Available at https://atvs.ii.uam.es/atvs/databases.jsp
ATVS- FakeFingerprint Database (ATVS-FFp DB)
Participation in the development and acquisition with the (then) PhD-student and co-worker J. Galbally (2250 real and associated fake fingerprint images). Fake samples captured from gummy fingers generated both with and without the cooperation of the user. Three different sensors used: flat-optical, flat-capacitive and sweeping-thermal.
Described in the paper: J. Galbally, F. Alonso-Fernandez, J. Fierrez, J. Ortega-Garcia, “A High Performance Fingerprint Liveness Detection Method Based on Quality Related Features”, Elsevier Future Generation Computer Systems Journal, 28 (1): 311-321, January 2012
Available at https://atvs.ii.uam.es/atvs/databases.jsp
Multimodal Biometric Database “BioSecure BMDB” (2007-2008)
Participation in the development, acquisition, supervision and distribution (600 individuals in 2 sessions at 11 European institutions) by coordinating the relevant WP under the public European R&D BioSecure FP6 NoE (IST-2002-507634)
Database used in various evaluations of technology (BMEC´07, BMFSC´09, BSEC´09)
Described in the paper: J. Ortega-Garcia, J. Fierrez, F. Alonso-Fernandez, et al. “The Multi-Scenario Multi-Environment BioSecure Multimodal Database (BMDB)”, IEEE Trans. On Pattern Analysis and Machine Intelligence, 32 (6): 1097-1111, June 2010, ISSN: 0162-8828
Available at https://biosecure.wp.imtbs-tsp.eu/biosecure-database/
Multimodal Biometric Database “BIOSECUR-ID” (2006)
Participation in the development, acquisition, supervision and distribution (400 individuals in 4 sessions at 6 Spanish institutions) under the public Spanish R&D project BIOSECUR-ID (TIC2003-08382-C05)
Described in the paper: J. Fierrez, J. Galbally, J. Ortega-Garcia, M. R. Freire, F. Alonso-Fernandez, D. Ramos, D. T. Toledano, J. Gonzalez-Rodriguez, J. A. Siguenza, J. Garrido-Salas, E. Anguiano, G. Gonzalez-de-Rivera, R. Ribalda, M. Faundez-Zanuy, J. A. Ortega, V. Cardeñoso-Payo, A. Viloria, C. E. Vivaracho, Q. I. Moro, J. J. Igarza, J. Sanchez, I. Hernaez, C. Orrite-Uruñuela, F. Martinez-Contreras and J. J. Gracia-Roche, “BiosecurID: A Multimodal Biometric Database", Pattern Analysis and Applications, 13 (2): 235-246, May 2010, ISSN: 1433-7541
Multimodal Biometric Database “BioSec” (2005-2006)
Participation in the development, acquisition, supervision and distribution (250 individuals in 4 sessions at 4 European institutions) under the public European R&D BioSec FP6 IP (IST-2002-001766)
Described in the paper: J. Fierrez, J. Ortega-Garcia, D. Torre-Toledano and J. Gonzalez-Rodriguez, "BioSec baseline corpus: A multimodal biometric database", Pattern Recognition, Vol. 40, n. 4, pp. 1389-1392, April 2007