Researchers have introduced MultiGait, a novel dataset and benchmark designed to assess the privacy risks associated with various smart city sensors. The dataset, which includes data from thermal, depth, and lidar cameras, captures gait information from 199 individuals across multiple sessions and perspectives. Initial benchmarks using state-of-the-art recognition systems reveal significant identity inference risks, even from sensors previously considered privacy-friendly, and highlight a gap in cross-session generalization for current methods. AI
IMPACT Highlights potential privacy vulnerabilities in emerging sensor technologies, necessitating further research into anonymization and secure data handling.
RANK_REASON The cluster contains a research paper introducing a new dataset and benchmark for privacy risk assessment. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
