Sign in the Air to Unlock: An Interface for authentication in Virtual and Augmented Reality Powered by Point-Voxel Cross-Attention Network

2026-07-01

AI Summary

Existing VR/AR authentication methods often compromise immersion, rely on external devices, or constrain natural user movements, making it challenging to balance security and usability. This work proposes a natural authentication interface based on mid-air handwritten signatures, introducing for the first time personalized and reproducible 3D air signatures into VR/AR environments. To effectively capture both local dynamic patterns and global spatial structure of signing trajectories, the authors design a Point-Voxel Cross-Attention Network (PV-Net). Experimental results demonstrate the method’s efficacy, achieving an equal error rate of 2.5% on the DeepAirSig dataset and 76% classification accuracy on the newly collected ImmAirSig dataset, thereby validating its practicality and robustness in real-world immersive scenarios.

Abstract

Significant advancement of immersive technologies such as Virtual and Augmented Reality (VR/AR) and their integration into diverse aspects of modern life need authentication interfaces that are secure, intuitive, and compatible with embodied interaction. Traditional methods such as passwords, PINs, and device-based logins, break immersion and rely on external hardware. Recent 3D-specific behavioral approaches, such as hand-gesture, eye-tracking, and electroencephalography (EEG)-based methods, offer promising alternatives but often require specialized sensors or constrain natural movement, limiting usability in dynamic environments. We present Sign in the Air to Unlock, an in-air signature interface that enables users to authenticate by signing naturally in 3D space which is a familiar, personal, and reproducible gesture. To realize this interface, we design a point-voxel Cross-Attention Network (PV-Net) that jointly models local motion dynamics and global spatial structure from 3D trajectories. The model is evaluated on two datasets: the public DeepAirSig dataset (1,800 signatures from 40 users) and ImmAirsig, a new dataset collected using Meta Quest 2 in immersive VR (880 samples from 22 users). PV-Net achieves an Equal Error Rate of 2.5% on DeepAirSig and 76% classification accuracy on ImmAirSig. These findings highlight the potential of 3D behavioral interfaces for seamless, user-centric authentication that merges security with natural interaction in immersive environments.

Problem

• authentication
• Virtual Reality
• Augmented Reality
• 3D behavioral interface
• immersive interaction

Innovation

• Point-Voxel Cross-Attention Network
• 3D behavioral authentication
• in-air signature
• immersive VR/AR
• PV-Net