iMARS is a research project funded by the European Commission, started in September 2020.
Official website: https://imars-project.eu/
The use of counterfeit travel documents at borders is a reality, which entails the risk of not identifying known criminals, including terrorists, or previously unknown subjects such as victims of human trafficking. Document fraud is one of the engines of organised crime in the European Union: although document fraud does not have a direct impact on most citizens in the European Union, it facilitates other serious and organised crime. It also has a strong impact on Members States since the financial consequences may be significant.
iMARS overall mission is to strengthen security at EU borders and improve operational capacity of passport application officers/border guards to assess ID documents validity against morphed images attacks during enrolment steps and at the border crossing stations.
iMARS will develop performant, affordable and customized morphing detection software capable for on the fly border processing but also investigate how image quality affects the morphing processes and the detection tools.ICPR_2022_CL_for_FM
The iMARS project will include all the steps and procedures that guide the work of the enrolment agencies and border control experts.
Software capabilities will be validated in real operational scenarios in several EU countries (with border to countries outside EU).
Publications:
-
- Borghi G., Graffieti G., Franco A., Maltoni D. Incremental Training of Face Morphing Detectors. Proceedings of the 26th International Conference of Pattern Recognition (ICPR), 2022 [PDF]
- Borghi G., Franco A., Graffieti G., Maltoni D. Automated Artifact Retouching in Morphed Images with Attention Maps. IEEE ACCESS, 2021 [PDF]
- Borghi G., Pancisi E., Ferrara M., Maltoni D. A Double Siamese Framework for Differential Morphing Attack Detection. MDPI Sensors, 2021 [PDF]