FPI-Det: a face--phone Interaction Dataset for phone-use detection and understanding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Jianqin, Wang, Tianqi, Zhang, Yu, Zhang, Yishu, Wang, Chenyuan, Dong, Allan, Wang, Zihao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912581814321152
author Gao, Jianqin
Wang, Tianqi
Zhang, Yu
Zhang, Yishu
Wang, Chenyuan
Dong, Allan
Wang, Zihao
author_facet Gao, Jianqin
Wang, Tianqi
Zhang, Yu
Zhang, Yishu
Wang, Chenyuan
Dong, Allan
Wang, Zihao
contents The widespread use of mobile devices has created new challenges for vision systems in safety monitoring, workplace productivity assessment, and attention management. Detecting whether a person is using a phone requires not only object recognition but also an understanding of behavioral context, which involves reasoning about the relationship between faces, hands, and devices under diverse conditions. Existing generic benchmarks do not fully capture such fine-grained human--device interactions. To address this gap, we introduce the FPI-Det, containing 22{,}879 images with synchronized annotations for faces and phones across workplace, education, transportation, and public scenarios. The dataset features extreme scale variation, frequent occlusions, and varied capture conditions. We evaluate representative YOLO and DETR detectors, providing baseline results and an analysis of performance across object sizes, occlusion levels, and environments. Source code and dataset is available at https://github.com/KvCgRv/FPI-Det.
format Preprint
id arxiv_https___arxiv_org_abs_2509_09111
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FPI-Det: a face--phone Interaction Dataset for phone-use detection and understanding
Gao, Jianqin
Wang, Tianqi
Zhang, Yu
Zhang, Yishu
Wang, Chenyuan
Dong, Allan
Wang, Zihao
Computer Vision and Pattern Recognition
The widespread use of mobile devices has created new challenges for vision systems in safety monitoring, workplace productivity assessment, and attention management. Detecting whether a person is using a phone requires not only object recognition but also an understanding of behavioral context, which involves reasoning about the relationship between faces, hands, and devices under diverse conditions. Existing generic benchmarks do not fully capture such fine-grained human--device interactions. To address this gap, we introduce the FPI-Det, containing 22{,}879 images with synchronized annotations for faces and phones across workplace, education, transportation, and public scenarios. The dataset features extreme scale variation, frequent occlusions, and varied capture conditions. We evaluate representative YOLO and DETR detectors, providing baseline results and an analysis of performance across object sizes, occlusion levels, and environments. Source code and dataset is available at https://github.com/KvCgRv/FPI-Det.
title FPI-Det: a face--phone Interaction Dataset for phone-use detection and understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.09111