Using unlabeled data in a sparse-coding framework for human activity recognition

Bhattacharya, Sourav; Nurmi, Petteri; Hammerla, Nils Y.; Plötz, Thomas

doi:10.1016/j.pmcj.2014.05.006

Cited by 104 publications

(78 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast to heuristic feature design, where domain specific expert knowledge is exploited to manually design features such as described above, the goal is to automatically discover meaningful representations of data. This is usually done by optimising an objective function that captures the appropriateness of the features, such as by energy minimisation or so-called "deep learning" (see [35] for a review) Building on this [36] developed sparse-coding framework for activity recognition exploits unlabelled sample data, whilst learning meaningful sparse feature representations. The authors give results on a benchmark dataset showing that their feature learning approach outperforms state-of-the-art approaches to analysing ADL, and claim that their approach will generalise well (see Section 2.7 for further discussion of this).…”

Section: Survey Of Feature Extraction Pipelinesmentioning

confidence: 99%

“…A related approach was taken by [36], using a sparse-coding framework for human activity recognition. In this case the authors used a clustering approach to group together sparse codes, rather than full CSC.…”

Section: Fixed Dictionariesmentioning

confidence: 99%

“…For all of the sparse coding techniques above, the coefficients that are learnt on each signal become the features for the classification algorithm, as proposed by [36]. We note that there has been some work in unifying dictionary learning and classification in a single optimisation framework [76], which has the potential to learn bases that are simultaneously useful for reconstruction and classification, we will leave this as a possible avenue for future work.…”

Section: Classification Using Sparse Codesmentioning

confidence: 99%

See 2 more Smart Citations

A Comprehensive Study of Activity Recognition Using Accelerometers

et al. 2018

View full text Add to dashboard Cite

This paper serves as a survey and empirical evaluation of the state-of-the-art in activity recognition methods using accelerometers. The paper is particularly focused on long-term activity recognition in real-world settings. In these environments, data collection is not a trivial matter; thus, there are performance trade-offs between prediction accuracy, which is not the sole system objective, and keeping the maintenance overhead at minimum levels. We examine research that has focused on the selection of activities, the features that are extracted from the accelerometer data, the segmentation of the time-series data, the locations of accelerometers, the selection and configuration trade-offs, the test/retest reliability, and the generalisation performance. Furthermore, we study these questions from an experimental platform and show, somewhat surprisingly, that many disparate experimental configurations yield comparable predictive performance on testing data. Our understanding of these results is that the experimental setup directly and indirectly defines a pathway for context to be delivered to the classifier, and that, in some settings, certain configurations are more optimal than alternatives. We conclude by identifying how the main results of this work can be used in practice, specifically in experimental configurations in challenging experimental conditions.

show abstract

Section: Survey Of Feature Extraction Pipelinesmentioning

confidence: 99%

Section: Fixed Dictionariesmentioning

confidence: 99%

Section: Classification Using Sparse Codesmentioning

confidence: 99%

See 1 more Smart Citation

A Comprehensive Study of Activity Recognition Using Accelerometers

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Plotz [11] used restricted Boltzmann machines to extract features. Li [12] and Bhattacharya [13] both used sparse-coding to extract features.…”

Section: Related Workmentioning

confidence: 99%

Recognition of Human Activities by Smartphone Sensors Using LSTM Neural Network

Zhao

Hou²,

Ma³

2018

Proceedings of the 2018 2nd International Conference on Artificial Intelligence: Technologies and Applications (ICAITA 2018)

View full text Add to dashboard Cite

Abstract-Human activities have been a hot research field. Many sensors are embedded in the smartphone, which makes mobile sensor become available. Sensors of smartphone can early get the human activities information, which can analyze the human behaviors and provide the useful information to the human. In this paper, we propose a new method to recognize the human activities, which is based on the LSTM neural network to extract features and classify using accelerometer sensor data and gyroscope sensor data. Experimental results show that using LSTM neural network and TensorFlow deep learning open source architecture to extract motion state characteristics, this method achieves human activities classification with an accuracy of up to 90.4%.

show abstract

“…location, time, work and other commitments) and addressing cost concerns. Furthermore, providing patients with marker of performance and ensuring a correct understanding of the content of rehabilitation will keep them motivated while supporting self-management [10], [11].…”

Section: Introductionmentioning

confidence: 99%

Estimation of Walking rate in Complex activity recognition

Kashanian¹,

Sharif²,

Akildyz³

2016

IJCATR

View full text Add to dashboard Cite

Physical activity recognition using embedded sensors has enabled by many context-aware applications in different areas. In sequential acceleration data there is a natural dependence between observations of movement or behavior, a fact that has been largely ignored in most analyses. In this paper, investigate the role that smart devices, including smartphones, can play in identifying activities of daily living. Monitoring and precisely quantifying users' physical activity with inertial measurement unit-based devices, for instance, has also proven to be important in health management of patients affected by chronic diseases, e.g. We show that their combination only improves the overall recognition performance when their individual performances are not very high, so that there is room for performance improvement. We show that the system can be used accurately to monitor both feet movement and use this result in many applications such as any playing. Time and frequency domain features of the signal were used to discriminate between activities, it demonstrates accuracy of 93% when employing a random forest analytical approach.

show abstract

Using unlabeled data in a sparse-coding framework for human activity recognition

Cited by 104 publications

References 44 publications

A Comprehensive Study of Activity Recognition Using Accelerometers

A Comprehensive Study of Activity Recognition Using Accelerometers

Recognition of Human Activities by Smartphone Sensors Using LSTM Neural Network

Estimation of Walking rate in Complex activity recognition

Contact Info

Product

Resources

About