Self-supervised Human Activity Recognition by Learning to Predict Cross-Dimensional Motion

TL;DR

This paper introduces a self-supervised learning approach for human activity recognition using smartphone accelerometer data, leveraging cross-dimensional motion prediction to learn effective representations that improve classification accuracy.

Contribution

The paper presents a novel cross-dimensional prediction scheme for self-supervised learning, enhancing feature extraction for human activity recognition from accelerometer data.

Findings

Outperforms existing methods on UCI HAR, MotionSense, and HAPT datasets.

Achieves state-of-the-art accuracy in human activity classification.

Demonstrates the effectiveness of cross-dimensional prediction in self-supervised learning.

Abstract

We propose the use of self-supervised learning for human activity recognition with smartphone accelerometer data. Our proposed solution consists of two steps. First, the representations of unlabeled input signals are learned by training a deep convolutional neural network to predict a segment of accelerometer values. Our model exploits a novel scheme to leverage past and present motion in x and y dimensions, as well as past values of the z axis to predict values in the z dimension. This cross-dimensional prediction approach results in effective pretext training with which our model learns to extract strong representations. Next, we freeze the convolution blocks and transfer the weights to our downstream network aimed at human activity recognition. For this task, we add a number of fully connected layers to the end of the frozen network and train the added layers with labeled accelerometer signals to learn to classify human activities. We evaluate the performance of our method on three publicly available human activity datasets: UCI HAR, MotionSense, and HAPT. The results show that our approach outperforms the existing methods and sets new state-of-the-art results.