VisGuardian: A Lightweight Group-based Privacy Control Technique For Front Camera Data From AR Glasses in Home Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Shuning, Zang, Qucheng, Hu, Yongquan `Owen', Du, Jiachen, Wang, Xueyang, Kong, Yan, Fu, Xinyi, Nanayakkara, Suranga, Yi, Xin, Li, Hewu
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917225187770368
author Zhang, Shuning
Zang, Qucheng
Hu, Yongquan `Owen'
Du, Jiachen
Wang, Xueyang
Kong, Yan
Fu, Xinyi
Nanayakkara, Suranga
Yi, Xin
Li, Hewu
author_facet Zhang, Shuning
Zang, Qucheng
Hu, Yongquan `Owen'
Du, Jiachen
Wang, Xueyang
Kong, Yan
Fu, Xinyi
Nanayakkara, Suranga
Yi, Xin
Li, Hewu
contents Always-on sensing of AI applications on AR glasses makes traditional permission techniques ill-suited for context-dependent visual data, especially within home environments. The home presents a highly challenging privacy context due to the high density of sensitive objects, and the frequent presence of non-consenting family members, and the intimate nature of daily routines, making it a critical focus area for scalable privacy control mechanisms. Existing fine-grained controls, while offering nuanced choices, are inefficient for managing multiple private objects. We propose VisGuardian, a fine-grained content-based visual permission technique for AR glasses. VisGuardian features a group-based control mechanism that enables users to efficiently manage permissions for multiple private objects. VisGuardian detects objects using YOLO and adopts a pre-classified schema to group them. By selecting a single object, users can efficiently obscure groups of related objects based on criteria including privacy sensitivity, object category, or spatial proximity. A technical evaluation shows VisGuardian achieves mAP50 of 0.6704 with only 14.0 ms latency and a 1.7% increase in battery consumption per hour. Furthermore, a user study (N=24) comparing VisGuardian to slider-based and object-based baselines found it to be significantly faster for setting permissions and was preferred by users for its efficiency, effectiveness, and ease of use.
format Preprint
id arxiv_https___arxiv_org_abs_2601_19502
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VisGuardian: A Lightweight Group-based Privacy Control Technique For Front Camera Data From AR Glasses in Home Environments
Zhang, Shuning
Zang, Qucheng
Hu, Yongquan `Owen'
Du, Jiachen
Wang, Xueyang
Kong, Yan
Fu, Xinyi
Nanayakkara, Suranga
Yi, Xin
Li, Hewu
Human-Computer Interaction
Cryptography and Security
Always-on sensing of AI applications on AR glasses makes traditional permission techniques ill-suited for context-dependent visual data, especially within home environments. The home presents a highly challenging privacy context due to the high density of sensitive objects, and the frequent presence of non-consenting family members, and the intimate nature of daily routines, making it a critical focus area for scalable privacy control mechanisms. Existing fine-grained controls, while offering nuanced choices, are inefficient for managing multiple private objects. We propose VisGuardian, a fine-grained content-based visual permission technique for AR glasses. VisGuardian features a group-based control mechanism that enables users to efficiently manage permissions for multiple private objects. VisGuardian detects objects using YOLO and adopts a pre-classified schema to group them. By selecting a single object, users can efficiently obscure groups of related objects based on criteria including privacy sensitivity, object category, or spatial proximity. A technical evaluation shows VisGuardian achieves mAP50 of 0.6704 with only 14.0 ms latency and a 1.7% increase in battery consumption per hour. Furthermore, a user study (N=24) comparing VisGuardian to slider-based and object-based baselines found it to be significantly faster for setting permissions and was preferred by users for its efficiency, effectiveness, and ease of use.
title VisGuardian: A Lightweight Group-based Privacy Control Technique For Front Camera Data From AR Glasses in Home Environments
topic Human-Computer Interaction
Cryptography and Security
url https://arxiv.org/abs/2601.19502