12 Sep Highlights from the 30th European Signal Processing Conference EUSPICO 2022
UVIGO has led together with CNRS, RMA, and UTT a special session proposal that was accepted for the 30th European Signal Processing Conference, EUSIPCO 2022, held in Belgrade (Serbia) last week. EUSIPCO is the premier European conference on signal processing and the main goal of the proposed special session, entitled “Source Heterogeneity Problems in Multimedia Forensics and Steganalysis”, was to attract papers investigating different ways to mitigate the significant drop in performance that occur in steganalysis and forensics when dealing with multimedia contents whose provenance is unknown. This problem, known as CSM (i.e., Cover-Source Mismatch) in steganalysis, often happens in practice and becomes especially relevant when it comes to tools based on machine learning techniques (where the contents composing the training set are not sufficiently diverse to represent the real world), but also when striving towards deploying research tools in an operational context, which is the ultimate goal of the UNCOVER project.
The special session received a total of 6 papers: 5 of them were finally accepted, of which, 4 were written by partners of the UNCOVER project (in particular: one paper from UVIGO, one paper from UIBK, one paper from CTU, and a joint paper by UTT, CNRS, and RMA). All works were presented at the conference by representatives of the partners, where fruitful discussions have taken place and will definitely favor the development of more robust tools within the UNCOVER project, so that in the near future they could be deployed as operational services to be used in practice by LEAs, forensics experts, etc. in their investigations.
The program of the conference: here
Some pictures of our representatives/posters:
Please find attached the prepared Video for the paper submission on YouTube and a link to the published paper on the research group page of Universität Innsbruck:
https://informationsecurity.uibk.ac.at/pdfs/BHB2022_EUSIPCO.pdf