ViFiCon: Vision and Wireless Association Via Self-Supervised Contrastive Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Meegan, Nicholas, Liu, Hansi, Cao, Bryan Bo, Alali, Abrar, Dana, Kristin, Gruteser, Marco, Jain, Shubham, Ashok, Ashwin
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915952358064128
author Meegan, Nicholas
Liu, Hansi
Cao, Bryan Bo
Alali, Abrar
Dana, Kristin
Gruteser, Marco
Jain, Shubham
Ashok, Ashwin
author_facet Meegan, Nicholas
Liu, Hansi
Cao, Bryan Bo
Alali, Abrar
Dana, Kristin
Gruteser, Marco
Jain, Shubham
Ashok, Ashwin
contents We introduce ViFiCon, a self-supervised contrastive scheme which learns a cross-modal association between vision and wireless modalities. Specifically, the system uses pedestrian data collected from RGB-D camera footage and WiFi Fine Time Measurements (FTM) from a user's smartphone device. Depth data from RGB-D (vision domain) is inherently linked with an observable pedestrian, but FTM data (wireless domain) is associated only to a smartphone on the network. We represent temporal sequences from both vision and wireless domains by stacking multi-person depth data sequences within an image representation. This simplicity allows both scene-wide processing and fewer vision and wireless features, alleviating privacy and energy associated with transmitting IMU data. To facilitate self-supervised learning, we design a scene-wide synchronization pretext task for our network and then employ the learned representation for the downstream multimodal association task. We show that compared to fully supervised state-of-the-art models, ViFiCon achieves high performance vision-to-wireless association of 92.63% in 25 frames sliding window fashion (2.5s), finding which bounding box corresponds to which smartphone device, without hand-labeled association examples for training data. Extensive experimental results demonstrate ViFiCon applicability in real-world systems when wireless data annotations are scarce.
format Preprint
id arxiv_https___arxiv_org_abs_2210_05513
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle ViFiCon: Vision and Wireless Association Via Self-Supervised Contrastive Learning
Meegan, Nicholas
Liu, Hansi
Cao, Bryan Bo
Alali, Abrar
Dana, Kristin
Gruteser, Marco
Jain, Shubham
Ashok, Ashwin
Computer Vision and Pattern Recognition
65D19, 68T45
I.4.0; I.4.10; I.5.4
We introduce ViFiCon, a self-supervised contrastive scheme which learns a cross-modal association between vision and wireless modalities. Specifically, the system uses pedestrian data collected from RGB-D camera footage and WiFi Fine Time Measurements (FTM) from a user's smartphone device. Depth data from RGB-D (vision domain) is inherently linked with an observable pedestrian, but FTM data (wireless domain) is associated only to a smartphone on the network. We represent temporal sequences from both vision and wireless domains by stacking multi-person depth data sequences within an image representation. This simplicity allows both scene-wide processing and fewer vision and wireless features, alleviating privacy and energy associated with transmitting IMU data. To facilitate self-supervised learning, we design a scene-wide synchronization pretext task for our network and then employ the learned representation for the downstream multimodal association task. We show that compared to fully supervised state-of-the-art models, ViFiCon achieves high performance vision-to-wireless association of 92.63% in 25 frames sliding window fashion (2.5s), finding which bounding box corresponds to which smartphone device, without hand-labeled association examples for training data. Extensive experimental results demonstrate ViFiCon applicability in real-world systems when wireless data annotations are scarce.
title ViFiCon: Vision and Wireless Association Via Self-Supervised Contrastive Learning
topic Computer Vision and Pattern Recognition
65D19, 68T45
I.4.0; I.4.10; I.5.4
url https://arxiv.org/abs/2210.05513