cs.LGMar 4, 2026

Activity Recognition from Smart Insole Sensor Data Using a Circular Dilated CNN

Authors: Yanhua Zhao

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Smart insoles equipped with pressure sensors, accelerometers, and gyroscopes offer a non-intrusive means of monitoring human gait and posture. We present an activity classification system based on a circular dilated convolutional neural network (CDCNN) that processes multi-modal time-series data from such insoles. The model operates on 160-frame windows with 24 channels (18 pressure, 3 accelerometer, 3 gyroscope axes), achieving 86.42% test accuracy in a subject-independent evaluation on a four-class task (Standing, Walking, Sitting, Tandem), compared with 87.83% for an extreme gradient-boosted tree (XGBoost) model trained on flattened data. Permutation feature importance reveals that inertial sensors (accelerometer and gyroscope) contribute substantially to discrimination. The approach is suitable for embedded deployment and real-time inference.

Figures & tables

Fig. 1: Class distribution of the compiled smart insole dataset.
Fig. 2: Example tri-axial accelerometer window for a single activity segment.
Fig. 3: Confusion matrix for CDCNN model on the held-out test set.
Subject IDSittingStandingTandemWalkingTotal
@BTrule []104040426830
1003920412804
1204300358788
1303580360718
143963960147939
155663985904061960
Tab. 1: Number of samples per subject and class.
SplitSubject IDs# samples
@BTrule []Train1, 10, 12, 14, 15, 18, 19, 23, 30, 31, 3214,047
Val17, 223,333
Test13, 16, 243,689
Tab. 2: Subject-wise train/validation/test split and sample counts.
Fig. 4: Permutation feature importance by channel over the test set.
Fig. 5: Permutation feature importance by channel over the test set for the XGBoost.

Explore similar work

CardsList

  1. Smart Insole Human Activity Recognition for Continuous Monitoring in Elderly Care

    Sep 16, 2026Edwin Rios, Antony Garcia, Fengpei Yuan +1Human Activity RecognitionWearable Sensing

    Department of Electrical and Computer Engineering Worcester Polytechnic Institute Worcester, MA, USA · Facultad de Ingenier´ıa El´ectrica Universidad Tecnol´ogica de Panam´a Panama City, Panama · Department of Robotics Engineering Worcester Polytechnic Institute Worcester, MA, USA

  2. An Open-Source, Open Data Approach to Activity Classification from Triaxial Accelerometry in an Ambulatory Setting

    Apr 10, 2026Sepideh Nikookar, Edward Tian, Harrison Hoffman +7Human Activity RecognitionWearable Health Monitoring

    Department of Biomedical Informatics, Emory University, Atlanta, GA, USA · LifeBell AI LLC, Atlanta, GA, USA · Department of Neurology, Emory University, Atlanta, GA, USA +3

  3. Robust Real-Time Monocular Human Pose Estimation with Insole-Based Sensor Fusion

    Oct 14, 2026Liang Yu, Tao Zhu, Xiangqing Lu +2Human Pose EstimationMultimodal Sensor Fusion

    University of South China, Hengyang, Hunan, China · Harbin Institute of Technology, Harbin, China · Dalian University of Technology, Dalian, China