Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSensor Data Augmentation with Resampling for Contrastive Learning in Human Activity Recognition
86
0
0.0
(
0
)
Added by
Tao Zhu
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Human activity recognition plays an increasingly important role not only in our daily lives, but also in the medical and rehabilitation fields. The development of deep learning has also contributed to the advancement of human activity recognition, but the large amount of data annotation work required to train deep learning models is a major obstacle to the development of human activity recognition. Contrastive learning has started to be used in the field of sensor-based human activity recognition due to its ability to avoid the cost of labeling large datasets and its ability to better distinguish between sample representations of different instances. Among them, data augmentation, an important part of contrast learning, has a significant impact on model effectiveness, but current data augmentation methods do not perform too successfully in contrast learning frameworks for wearable sensor-based activity recognition. To optimize the effect of contrast learning models, in this paper, we investigate the sampling frequency of sensors and propose a resampling data augmentation method. In addition, we also propose a contrast learning framework based on human activity recognition and apply the resampling augmentation method to the data augmentation phase of contrast learning. The experimental results show that the resampling augmentation method outperforms supervised learning by 9.88% on UCI HAR and 7.69% on Motion Sensor in the fine-tuning evaluation of contrast learning with a small amount of labeled data, and also reveal that not all data augmentation methods will have positive effects in the contrast learning framework. Finally, we explored the influence of the combination of different augmentation methods on contrastive learning, and the experimental results showed that the effect of most combination augmentation methods was better than that of single augmentation.
rate research
Read More
Data augmentation is a widely used technique in classification to increase data used in training. It improves generalization and reduces amount of annotated human activity data needed for training which reduces labour and time needed with the dataset. Sensor time-series data, unlike images, cannot be augmented by computationally simple transformation algorithms. State of the art models like Recurrent Generative Adversarial Networks (RGAN) are used to generate realistic synthetic data. In this paper, transformer based generative adversarial networks which have global attention on data, are compared on PAMAP2 and Real World Human Activity Recognition data sets with RGAN. The newer approach provides improvements in time and savings in computational resources needed for data augmentation than previous approach.
The vast proliferation of sensor devices and Internet of Things enables the applications of sensor-based activity recognition. However, there exist substantial challenges that could influence the performance of the recognition system in practical scenarios. Recently, as deep learning has demonstrated its effectiveness in many areas, plenty of deep methods have been investigated to address the challenges in activity recognition. In this study, we present a survey of the state-of-the-art deep learning methods for sensor-based human activity recognition. We first introduce the multi-modality of the sensory data and provide information for public datasets that can be used for evaluation in different challenge tasks. We then propose a new taxonomy to structure the deep methods by challenges. Challenges and challenge-related deep methods are summarized and analyzed to form an overview of the current research progress. At the end of this work, we discuss the open issues and provide some insights for future directions.
Wearable sensor-based human activity recognition (HAR) has been a research focus in the field of ubiquitous and mobile computing for years. In recent years, many deep models have been applied to HAR problems. However, deep learning methods typically require a large amount of data for models to generalize well. Significant variances caused by different participants or diverse sensor devices limit the direct application of a pre-trained model to a subject or device that has not been seen before. To address these problems, we present an invariant feature learning framework (IFLF) that extracts common information shared across subjects and devices. IFLF incorporates two learning paradigms: 1) meta-learning to capture robust features across seen domains and adapt to an unseen one with similarity-based data selection; 2) multi-task learning to deal with data shortage and enhance overall performance via knowledge sharing among different subjects. Experiments demonstrated that IFLF is effective in handling both subject and device diversion across popular open datasets and an in-house dataset. It outperforms a baseline model of up to 40% in test accuracy.
Recent years have witnessed the rapid development of human activity recognition (HAR) based on wearable sensor data. One can find many practical applications in this area, especially in the field of health care. Many machine learning algorithms such as Decision Trees, Support Vector Machine, Naive Bayes, K-Nearest Neighbor, and Multilayer Perceptron are successfully used in HAR. Although these methods are fast and easy for implementation, they still have some limitations due to poor performance in a number of situations. In this paper, we propose a novel method based on the ensemble learning to boost the performance of these machine learning methods for HAR.
Human Activity Recognition from body-worn sensor data poses an inherent challenge in capturing spatial and temporal dependencies of time-series signals. In this regard, the existing recurrent or convolutional or their hybrid models for activity recognition struggle to capture spatio-temporal context from the feature space of sensor reading sequence. To address this complex problem, we propose a self-attention based neural network model that foregoes recurrent architectures and utilizes different types of attention mechanisms to generate higher dimensional feature representation used for classification. We performed extensive experiments on four popular publicly available HAR datasets: PAMAP2, Opportunity, Skoda and USC-HAD. Our model achieve significant performance improvement over recent state-of-the-art models in both benchmark test subjects and Leave-one-subject-out evaluation. We also observe that the sensor attention maps produced by our model is able capture the importance of the modality and placement of the sensors in predicting the different activity classes.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
