Saved in:
Bibliographic Details
Main Authors: Yang, Xuefeng, Zhang, Shiheng, Guan, Jian, Xiao, Feiyang, Lu, Wei, Zhu, Qiaoxi
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.19404
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909603262889984
author Yang, Xuefeng
Zhang, Shiheng
Guan, Jian
Xiao, Feiyang
Lu, Wei
Zhu, Qiaoxi
author_facet Yang, Xuefeng
Zhang, Shiheng
Guan, Jian
Xiao, Feiyang
Lu, Wei
Zhu, Qiaoxi
contents This study is based on the ICASSP 2025 Signal Processing Grand Challenge's Accelerometer-Based Person-in-Bed Detection Challenge, which aims to determine bed occupancy using accelerometer signals. The task is divided into two tracks: "in bed" and "not in bed" segmented detection, and streaming detection, facing challenges such as individual differences, posture variations, and external disturbances. We propose a spectral-temporal fusion-based feature representation method with mixup data augmentation, and adopt Intersection over Union (IoU) loss to optimize detection accuracy. In the two tracks, our method achieved outstanding results of 100.00% and 95.55% in detection scores, securing first place and third place, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2412_19404
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Spectral-Temporal Fusion Representation for Person-in-Bed Detection
Yang, Xuefeng
Zhang, Shiheng
Guan, Jian
Xiao, Feiyang
Lu, Wei
Zhu, Qiaoxi
Signal Processing
Computer Vision and Pattern Recognition
Machine Learning
This study is based on the ICASSP 2025 Signal Processing Grand Challenge's Accelerometer-Based Person-in-Bed Detection Challenge, which aims to determine bed occupancy using accelerometer signals. The task is divided into two tracks: "in bed" and "not in bed" segmented detection, and streaming detection, facing challenges such as individual differences, posture variations, and external disturbances. We propose a spectral-temporal fusion-based feature representation method with mixup data augmentation, and adopt Intersection over Union (IoU) loss to optimize detection accuracy. In the two tracks, our method achieved outstanding results of 100.00% and 95.55% in detection scores, securing first place and third place, respectively.
title Spectral-Temporal Fusion Representation for Person-in-Bed Detection
topic Signal Processing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2412.19404