Similarity Embedding Networks for Robust Human Activity Recognition

TL;DR

This paper introduces a similarity embedding neural network for human activity recognition that improves robustness to noisy labels and small datasets, outperforming existing models in real-world scenarios.

Contribution

The paper presents a novel similarity embedding network trained with pairwise similarity loss, enhancing HAR performance and robustness to label noise and limited data.

Findings

Outperforms state-of-the-art models on public HAR datasets

Robust to mislabeled training data and noise

Effective in small dataset scenarios

Abstract

Deep learning models for human activity recognition (HAR) based on sensor data have been heavily studied recently. However, the generalization ability of deep models on complex real-world HAR data is limited by the availability of high-quality labeled activity data, which are hard to obtain. In this paper, we design a similarity embedding neural network that maps input sensor signals onto real vectors through carefully designed convolutional and LSTM layers. The embedding network is trained with a pairwise similarity loss, encouraging the clustering of samples from the same class in the embedded real space, and can be effectively trained on a small dataset and even on a noisy dataset with mislabeled samples. Based on the learned embeddings, we further propose both nonparametric and parametric approaches for activity recognition. Extensive evaluation based on two public datasets has shown that the proposed similarity embedding network significantly outperforms state-of-the-art deep models on HAR classification tasks, is robust to mislabeled samples in the training set, and can also be used to effectively denoise a noisy dataset.