Deep Architectures for Human Activity Recognition using Sensors

Baloch, Zartasha; Shaikh, Faisal Karim; Unar, Mukhtiar Ali

doi:10.17993/3ctecno.2019.specialissue2.14-35

Cited by 8 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the field of sensor-based HAR, deep learning methods recently gained a lot of attention. According to Baloch et al [ 42 ], the most common neural network architectures for sensor based activity recognition are CNNs (40%), Recurrent Neural Nets (RNNs) including LSTMs (30%) and hybrid models as convolutional LSTM models (15%) For each of this network types we implemented a basic architecture for the shot and pass classification in football. From the kinetic and kinematic analysis of side-foot and instep kicks in football we know that the two kicking techniques differ in foot speed and foot rotation.…”

Section: Methodsmentioning

confidence: 99%

From the Laboratory to the Field: IMU-Based Shot and Pass Detection in Football Training and Game Scenarios Using Deep Learning

Stoeve

Schuldhaus

Gamp

et al. 2021

Sensors

View full text Add to dashboard Cite

The applicability of sensor-based human activity recognition in sports has been repeatedly shown for laboratory settings. However, the transferability to real-world scenarios cannot be granted due to limitations on data and evaluation methods. On the example of football shot and pass detection against a null class we explore the influence of those factors for real-world event classification in field sports. For this purpose we compare the performance of an established Support Vector Machine (SVM) for laboratory settings from literature to the performance in three evaluation scenarios gradually evolving from laboratory settings to real-world scenarios. In addition, three different types of neural networks, namely a convolutional neural net (CNN), a long short term memory net (LSTM) and a convolutional LSTM (convLSTM) are compared. Results indicate that the SVM is not able to reliably solve the investigated three-class problem. In contrast, all deep learning models reach high classification scores showing the general feasibility of event detection in real-world sports scenarios using deep learning. The maximum performance with a weighted f1-score of 0.93 was reported by the CNN. The study provides valuable insights for sports assessment under practically relevant conditions. In particular, it shows that (1) the discriminative power of established features needs to be reevaluated when real-world conditions are assessed, (2) the selection of an appropriate dataset and evaluation method are both required to evaluate real-world applicability and (3) deep learning-based methods yield promising results for real-world HAR in sports despite high variations in the execution of activities.

show abstract

Section: Methodsmentioning

confidence: 99%

From the Laboratory to the Field: IMU-Based Shot and Pass Detection in Football Training and Game Scenarios Using Deep Learning

Stoeve

Schuldhaus

Gamp

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…While many quality datasets are available for benchmarking AR algorithms [43], we used the Opportunity Activity Recognition dataset to demonstrate FilterNet due to its wide usage as a benchmark dataset in the activity recognition field [7], its relatively large size (about 6 h of recordings), and its diverse array of sensors and labels (from which we can choose various subsets). Furthermore, like the real-world datasets for which FilterNet was developed, much of the Opportunity dataset consists of the null class-that is, regions without labeled behaviors.…”

Section: Benchmark Datasetmentioning

confidence: 99%

FilterNet: A Many-to-Many Deep Learning Architecture for Time Series Classification

Chambers¹,

Yoder²

2020

Sensors

View full text Add to dashboard Cite

In this paper, we present and benchmark FilterNet, a flexible deep learning architecture for time series classification tasks, such as activity recognition via multichannel sensor data. It adapts popular convolutional neural network (CNN) and long short-term memory (LSTM) motifs which have excelled in activity recognition benchmarks, implementing them in a many-to-many architecture to markedly improve frame-by-frame accuracy, event segmentation accuracy, model size, and computational efficiency. We propose several model variants, evaluate them alongside other published models using the Opportunity benchmark dataset, demonstrate the effect of model ensembling and of altering key parameters, and quantify the quality of the models’ segmentation of discrete events. We also offer recommendations for use and suggest potential model extensions. FilterNet advances the state of the art in all measured accuracy and speed metrics when applied to the benchmarked dataset, and it can be extensively customized for other applications.

show abstract

“…While many quality datasets are available for benchmarking AR algorithms [43], we use the Opportunity Activity Recognition dataset to demonstrate FilterNet due to its wide usage as a benchmark dataset in the activity recognition field [7], its relatively large size (about 6 hours of recordings), and its diverse array of sensors and labels (from which we can choose various subsets). Furthermore, like the real-world datasets for which FilterNet was developed, much of the Opportunity dataset consists of the null class -that is, regions without labeled behaviors.…”

Section: Benchmark Datasetmentioning

confidence: 99%

FilterNet: A Many-to-Many Deep Learning Architecture for Time Series Classification

Chambers¹,

Yoder²

2020

Preprint

View full text Add to dashboard Cite

We present and benchmark FilterNet, a flexible deep learning architecture for time series classification tasks, such as activity recognition via multichannel sensor data. It adapts popular CNN and CNN-LSTM motifs which have excelled in activity recognition benchmarks, implementing them in a many-to-many architecture to markedly improve frame-by-frame accuracy, event segmentation accuracy, model size, and computational efficiency. We propose several model variants, evaluate them alongside other published models using the Opportunity benchmark dataset, demonstrate the effect of model ensembling and of altering key parameters, and quantify the quality of the models' segmentation of discrete events. We also offer recommendations for use and suggest potential model extensions. FilterNet advances the state of the art in all measured accuracy and speed metrics on the benchmarked dataset, and it can be extensively customized for other applications.

show abstract

Deep Architectures for Human Activity Recognition using Sensors

Cited by 8 publications

References 34 publications

From the Laboratory to the Field: IMU-Based Shot and Pass Detection in Football Training and Game Scenarios Using Deep Learning

From the Laboratory to the Field: IMU-Based Shot and Pass Detection in Football Training and Game Scenarios Using Deep Learning

FilterNet: A Many-to-Many Deep Learning Architecture for Time Series Classification

FilterNet: A Many-to-Many Deep Learning Architecture for Time Series Classification

Contact Info

Product

Resources

About