On the role of features in human activity recognition

Haresamudram, Harish; Anderson, David V.; Plötz, Thomas

doi:10.1145/3341163.3347727

Cited by 63 publications

(56 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate our approach, we consider 7 of the most widely recognized datasets in HAR [28], comparing for each dataset the performance of our method against the methods in the literature that we found have achieved highest performance. We provide the comparison with respect to at most 5 other approaches with best performance in the literature for each dataset.…”

Section: Evaluation Methodologymentioning

confidence: 99%

Improving Wearable-Based Activity Recognition Using Image Representations

Guinea

Sarabchian

Mühlhäuser

2022

Sensors

View full text Add to dashboard Cite

Activity recognition based on inertial sensors is an essential task in mobile and ubiquitous computing. To date, the best performing approaches in this task are based on deep learning models. Although the performance of the approaches has been increasingly improving, a number of issues still remain. Specifically, in this paper we focus on the issue of the dependence of today’s state-of-the-art approaches to complex ad hoc deep learning convolutional neural networks (CNNs), recurrent neural networks (RNNs), or a combination of both, which require specialized knowledge and considerable effort for their construction and optimal tuning. To address this issue, in this paper we propose an approach that automatically transforms the inertial sensors time-series data into images that represent in pixel form patterns found over time, allowing even a simple CNN to outperform complex ad hoc deep learning models that combine RNNs and CNNs for activity recognition. We conducted an extensive evaluation considering seven benchmark datasets that are among the most relevant in activity recognition. Our results demonstrate that our approach is able to outperform the state of the art in all cases, based on image representations that are generated through a process that is easy to implement, modify, and extend further, without the need of developing complex deep learning models.

show abstract

Section: Evaluation Methodologymentioning

confidence: 99%

Improving Wearable-Based Activity Recognition Using Image Representations

Guinea

Sarabchian

Mühlhäuser

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Plötz et al [6] demonstrated that feature learning using deep belief networks (DBN) can lead to significantly improved and especially generalized classification performance downstream when compared to hand-crafted features. Other variants of feature learning utilized more contemporary forms of autoencoder networks showing similar improvements in feature generalization capabilities [23,24].…”

Section: Feature Learningmentioning

confidence: 99%

“…Relaxing the requirements on annotations even further, one could explore to what extent virtual IMU data can be used for fully unsupervised learning scenarios where, for examples, feature representations are learned directly from raw sensor data [6,24].…”

Section: Virtual Imu Data As Basis For Alternatives To Supervised Learningmentioning

confidence: 99%

Complex Deep Neural Networks from Large Scale Virtual IMU Data for Effective Human Activity Recognition Using Wearables

Kwon

Abowd

Plötz

2021

Sensors

Self Cite

View full text Add to dashboard Cite

Supervised training of human activity recognition (HAR) systems based on body-worn inertial measurement units (IMUs) is often constrained by the typically rather small amounts of labeled sample data. Systems like IMUTube have been introduced that employ cross-modality transfer approaches to convert videos of activities of interest into virtual IMU data. We demonstrate for the first time how such large-scale virtual IMU datasets can be used to train HAR systems that are substantially more complex than the state-of-the-art. Complexity is thereby represented by the number of model parameters that can be trained robustly. Our models contain components that are dedicated to capture the essentials of IMU data as they are of relevance for activity recognition, which increased the number of trainable parameters by a factor of 1100 compared to state-of-the-art model architectures. We evaluate the new model architecture on the challenging task of analyzing free-weight gym exercises, specifically on classifying 13 dumbbell execises. We have collected around 41 h of virtual IMU data using IMUTube from exercise videos available from YouTube. The proposed model is trained with the large amount of virtual IMU data and calibrated with a mere 36 min of real IMU data. The trained model was evaluated on a real IMU dataset and we demonstrate the substantial performance improvements of 20% absolute F1 score compared to the state-of-the-art convolutional models in HAR.

show abstract

“…In [22] determining the optimal window length for each of the activities of interest in a given task shows substantially improved performance over using a standard window length across all activities. As shown in [16], feature engineering is an important aspect of learning representations that can be used to separate data belonging to different activity classes. Signal based statistical features were succeeded by more sophisticated feature extraction techniques such as in the case of ECDF features [14].…”

Section: Task Complexity Analysismentioning

confidence: 99%

Deriving Effective Human Activity Recognition Systems through Objective Task Complexity Assessment

Hiremath

Plötz

2020

Proc. ACM Interact. Mob. Wearable Ubiquitous Technol.

Self Cite

View full text Add to dashboard Cite

Research in sensor based human activity recognition (HAR) has been a core concern of the mobile and ubiquitous computing community. Sophisticated systems have been developed with the main view on applications of HAR methods in research settings. This work addresses a related yet practically different problem that mainly focuses on users of HAR technology. We acknowledge that practitioners from outside the core HAR research community are motivated to employ HAR methods for practical deployments. Even though standard processing approaches exist, arguably, often times substantial modifications are necessary to derive effective analysis systems. It is not always clear a-priori how challenging a HAR task actually is and what dimensions of an analysis pipeline are crucial for successful automated assessments. In practice this can lead to disappointing results or disproportionate efforts that have to be invested into the optimization of data analysis pipelines, that were supposed to work "out of the box". We present a framework for the objective complexity assessment of HAR tasks that directly supports practitioners' decision making of whether and how to employ HAR for their deployments. We map a HAR task onto a vectorial representation that allows us to analyse the inherent challenges of the task and to draw conclusions through similarity analysis with regards to existing tasks. We validate our complexity assessment framework on 23 HAR datasets and derive a data-driven categorization of human activity recognition. We demonstrate how our objective analysis can be used to inform the deployment of HAR systems in practical scenarios.

show abstract

On the role of features in human activity recognition

Cited by 63 publications

References 25 publications

Improving Wearable-Based Activity Recognition Using Image Representations

Improving Wearable-Based Activity Recognition Using Image Representations

Complex Deep Neural Networks from Large Scale Virtual IMU Data for Effective Human Activity Recognition Using Wearables

Deriving Effective Human Activity Recognition Systems through Objective Task Complexity Assessment

Contact Info

Product

Resources

About