Feature learning for Human Activity Recognition using Convolutional Neural Networks

Abstract

The use of Convolutional Neural Networks (CNNs) as a feature learning method for Human Activity Recognition (HAR) is becoming more and more common. Unlike conventional machine learning methods, which require domain-specific expertise, CNNs can extract features automatically. On the other hand, CNNs require a training phase, making them prone to the cold-start problem. In this work, a case study is presented where the use of a pre-trained CNN feature extractor is evaluated under realistic conditions. The case study consists of two main steps: (1) different topologies and parameters are assessed to identify the best candidate models for HAR, thus obtaining a pre-trained CNN model. The pre-trained model (2) is then employed as feature extractor evaluating its use with a large scale real-world dataset. Two CNN applications were considered: Inertial Measurement Unit (IMU) and audio based HAR. For the IMU data, balanced accuracy was 91.98% on the UCI-HAR dataset, and 67.51% on the real-world Extrasensory dataset. For the audio data, the balanced accuracy was 92.30% on the DCASE 2017 dataset, and 35.24% on the Extrasensory dataset.

Deep Sensing: Inertial and Ambient Sensing for Activity Context Recognition Using Deep Convolutional Neural Networks

“…[ 6 , 7 , 8 , 9 , 10 , 11 ]. DCNNs are now widely used for the development of automatic human activity context recognition [ 6 , 7 , 8 , 9 , 12 , 13 , 14 , 15 , 16 , 17 , 18 ]. Representational learning of activity context from raw sensor data using a DCNN has been proposed for automatic feature extraction in activity context recognition [ 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 ].…”

“…Besides, deep learning algorithms have the capability for unsupervised and incremental learning because of its deep network structure compared to the traditional neural network. A DCNN is composed of multiple building blocks, such as convolutional layers, pooling layers, and fully connected layers [ 9 , 13 , 14 , 15 , 16 , 17 ]. It has been designed to automatically and adaptively learn spatial hierarchies of features, from low to high-level patterns, through backpropagation algorithm [ 16 , 19 , 20 , 21 , 22 ].…”

Multi-Step Short-Term Wind Speed Prediction Using a Residual Dilated Causal Convolutional Network with Nonlinear Attention

“…In [27], it was shown that the CNN model could extract a suitable internal structure to produce deep features automatically and outperform state-of-the-art methods on real-world datasets. CNN architectures, unlike other feature-based models, do not require sophisticated feature engineering [28][29][30]. These papers provided supportive results and reviews showing that 1D CNN could be used as an automatic feature extraction mechanism for one-dimensional data.…”

Inertial Sensor-Based Instrumented Cane for Real-Time Walking Cane Kinematics Estimation

“…Unlike neural networks with a single hidden layer NN, CNN are able to learn how to extract more detailed features as its depth increases. Being based on the visual cortex, CNN models are commonly used in image-based applications, but have also been used in inertial sensor-sensor based applications such as activity recognition [ 31 ] and the detection of Parkinson’s disease-related events [ 32 , 33 ]. In such cases, instead of using an image input, the input is typically replaced by a 2D ( n × m ) matrix containing n samples from m inertial sensor signals.…”