Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
The paper introduces MultiGait, a comprehensive dataset designed to investigate the privacy risks associated with various smart city sensors, including thermal and depth cameras. By validating the dataset with state-of-the-art recognition systems, the authors highlight significant identity inference risks even from sensors typically considered privacy-friendly. The findings reveal critical gaps in cross-session generalization, emphasizing the need for further research in biometric privacy and sensor deployment strategies.
Even sensors thought to be privacy-friendly pose significant identity inference risks, challenging common assumptions in smart city deployments.
A lack of suitable datasets has limited the research into the privacy risks of novel smart city sensors, such as thermal cameras, depth cameras, and lidar. Given the number of unsubstantiated privacy claims and their potential widespread deployment into many people's everyday life, understanding the privacy risks of these sensors -- in isolation and in like-for-like comparisons -- is crucial. With MultiGait, we collected the first multi-sensor, multi-perspective, multi-session gait-focused dataset, for the corresponding, and additional more far-reaching investigations. The dataset, validated with multiple state-of-the-art recognition systems, comprises various walking modes and annotated personal attributes for 199 individuals, to ensure the benefit for advanced studies including cross-sensor recognition and anonymization at the edge. MultiGait represents a foundation for rigorous privacy investigations, demonstrated through an extensive identity inference benchmark across eight sensors, four perspectives, and three recording sessions. Our benchmark incidentally reveals that sensors often assumed to be privacy-friendly do still entail considerable identity inference risks, while the poor cross-session generalization of existing methods underscores an important research gap.