.png){kind=icon}
.png){kind=icon}
Return
Zero-Shot Face Authentication in Multi-User Scenarios Using mmWave Signals
.png){kind=icon}
PRE
AI
J
俞
H
Y
DOI：10.1109/tmc.2026.3698153.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Face authentication has become an increasingly attractive and prevalent technology in human-computer interaction. Recent works have explored radio frequency (RF) signals for illumination-robust and privacy-preserving face authentication. However, these approaches require all users to undergo compulsory prior registration and cannot effectively handle unregistered users, which restricts their practical applicability in real-world scenarios. In this paper, we present a <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mm</u>Wave-based <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m</u>ulti-user face authentication system, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i>, which performs zero-shot face authentication in complex multi-user scenarios using a commercial off-the-shelf (COTS) mmWave radar. First, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i> collects mmWave signals and reconstructs spatial mappings by generating 3D heatmaps. Based on the 3D heatmaps, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i> detects and separates faces in multi-user scenarios through template matching. Then, unique facial features are extracted and transformed to a hypersphere manifold, which reserves feature space for unregistered users. By introducing dummy classifiers in the authentication model, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i> optimizes decision boundaries between registered and unregistered users. Furthermore, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i> enhances its generalization ability for zero-shot authentication by incorporating pseudo unknowns during model training. Extensive experiments in real-world environments demonstrate that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">m<sup>3</sup>FacePass</i> achieves an authentication accuracy of 96.6% for registered users and 91.1% for unregistered users in multi-user scenarios.
Keywords:
MmWave signals
face authentication
multi-user scenarios
zero-shot learning
wireless sensing
Journal
IF:
9.2
Papers:
5.6K
Citations:
1.8W
Organization
Cited Papers