Few-Shot Scene Recognition, a New Proxy Task for Child Sexual Abuse Material Detection

• Grant number: 22/14690-8
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: April 01, 2023
• End date: October 30, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Sandra Eliza Fontes de Avila
• Grantee: Leo Sampaio Ferraz Ribeiro
• Host Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Associated research grant: 13/08293-7 - CCES - Center for Computational Engineering and Sciences, AP.CEPID

Abstract

The spread of Child Sexual Abuse Material (CSAM) over the internet is an ever-growing, worldwide problem that re-victimizes its victims and becomes increasingly intractable with current manual checking and investigative efforts from law enforcement agencies. Furthermore, the burden of dealing with this kind of data wears heavily on agents that have to categorize hundreds of terabytes of data every year in search of victims and perpetrators. There is then a clear need for automated CSAM recognition systems. In this project, we propose to design and prove the effectiveness of a new proxy task for CSAM detection that dispenses access to such sensitive data for parameter tuning. Few-shot Scene Recognition, our chosen task, should provide training and evaluation support for designing models that can then be given to specialized law-enforcement agents for CSAM detection, who are and should always be the only personnel with access to these materials. We highlight that our team is at the forefront of such research worldwide being responsible for groundbreaking results associated with Digital Forensics, Machine Learning, and Computer Vision.