The Devil Is in the Details: An Efficient Convolutional Neural Network for Transport Mode Detection

TL;DR

This paper presents a compact, optimized convolutional neural network for transport mode detection that achieves comparable accuracy to larger models while significantly reducing computational resources, and introduces improved preprocessing methods for variable-length signals.

Contribution

The authors develop a small, efficient CNN model for transport mode detection that rivals larger models and propose better preprocessing techniques for handling signals of different lengths.

Findings

Small models with tens of thousands of parameters perform as well as larger models.

Optimized preprocessing improves signal length handling.

Lighter CNNs can replace heavier RNNs for this task.

Abstract

Transport mode detection is a classification problem aiming to design an algorithm that can infer the transport mode of a user given multimodal signals (GPS and/or inertial sensors). It has many applications, such as carbon footprint tracking, mobility behaviour analysis, or real-time door-to-door smart planning. Most current approaches rely on a classification step using Machine Learning techniques, and, like in many other classification problems, deep learning approaches usually achieve better results than traditional machine learning ones using handcrafted features. Deep models, however, have a notable downside: they are usually heavy, both in terms of memory space and processing cost. We show that a small, optimized model can perform as well as a current deep model. During our experiments on the GeoLife and SHL 2018 datasets, we obtain models with tens of thousands of parameters, that is, 10 to 1,000 times less parameters and operations than networks from the state of the art, which still reach a comparable performance. We also show, using the aforementioned datasets, that the current preprocessing used to deal with signals of different lengths is suboptimal, and we provide better replacements. Finally, we introduce a way to use signals with different lengths with the lighter Convolutional neural networks, without using the heavier Recurrent Neural Networks.