No abstract

No abstract

Many human activities consist of physical gestures that tend to be performed in certain sequences. Wearable inertial sensor data have as a consequence been employed to automatically detect human activities, lately predominantly with deep learning methods. This article focuses on the necessity of recurrent layers—more specifically Long Short-Term Memory (LSTM) layers—in common Deep Learning architectures for Human Activity Recognition (HAR). Our experimental pipeline investigates the effects of employing none, one, or two LSTM layers, as well as different layers' sizes, within the popular DeepConvLSTM architecture. We evaluate the architecture's performance on five well-known activity recognition datasets and provide an in-depth analysis of the per-class results, showing trends which type of activities or datasets profit the most from the removal of LSTM layers. For 4 out of 5 datasets, an altered architecture with one LSTM layer produces the best prediction results. In our previous work we already investigated the impact of a 2-layered LSTM when dealing with sequential activity data. Extending upon this, we now propose a metric, rGP, which aims to measure the effectiveness of learned temporal patterns for a dataset and can be used as a decision metric whether to include recurrent layers into a network at all. Even for datasets including activities without explicit temporal processes, the rGP can be high, suggesting that temporal patterns were learned, and consequently convolutional networks are being outperformed by networks including recurrent layers. We conclude this article by putting forward the question to what degree popular HAR datasets contain unwanted temporal dependencies, which if not taken care of, can benefit networks in achieving high benchmark scores and give a false sense of overall generability to a real-world setting.

No abstract

No abstract

Purpose: Inertial-based trackers have become a common tool in data capture for ambulatory studies that aim at characterizing physical activity. Many systems that perform remote recording of accelerometer data use commercial trackers and black-box aggregation algorithms, often resulting in data that are locked into proprietary formats and metrics that make later replication or comparison difficult.Methods: The primary purpose of this manuscript is to validate an open-source ambulatory assessment system that consists of hardware devices, algorithms, and software components of our approach. We report on two validation experiments, one lab-based treadmill study on a convenience sample of 16 volunteers and one ’in vivo’ study with 28 volunteers suffering from diabetes or cardiovascular disease.Results: A comparison between data from ActiGraph GT9X trackers and our proposed system reveals that the original inertial sensor signals at the wrist strongly correlate (Pearson correlation coefficients for raw inertial sensor signals of 0.97 in the controlled treadmill-walking setting) and that estimated steps from an open-source wrist-based detection approach correlate with the hip-worn ActiGraph output (average Pearson correlation coefficients of 0.81 for minute-wise comparisons of detected steps) in day-long ambulatory data.Conclusion: Recording inertial sensor data in a standardized form and relying on open-source algorithms on these data form a promising methodology that ensures that datasets can be replicated or enriched long after the wearable trackers have been decommissioned.

We present a benchmark dataset for evaluating physical human activity recognition methods from wrist-worn sensors, for the specific setting of basketball training, drills, and games. Basketball activities lend themselves well for measurement by wrist-worn inertial sensors, and systems that are able to detect such sport-relevant activities could be used in applications of game analysis, guided training, and personal physical activity tracking. The dataset was recorded from two teams in separate countries (USA and Germany) with a total of 24 players who wore an inertial sensor on their wrist, during both a repetitive basketball training session and a game. Particular features of this dataset include an inherent variance through cultural differences in game rules and styles as the data was recorded in two countries, as well as different sport skill levels since the participants were heterogeneous in terms of prior basketball experience. We illustrate the dataset’s features in several time-series analyses and report on a baseline classification performance study with two state-of-the-art deep learning architectures.

Recent studies in Human Activity Recognition (HAR) have shown that Deep Learning methods are able to outperform classical Machine Learning algorithms. One popular Deep Learning architecture in HAR is the DeepConvLSTM. In this paper we propose to alter the DeepConvLSTM architecture to employ a 1-layered instead of a 2-layered LSTM. We validate our architecture change on 5 publicly available HAR datasets by comparing the predictive performance with and without the change employing varying hidden units within the LSTM layer(s). Results show that across all datasets, our architecture consistently improves on the original one: Recognition performance increases up to 11.7% for the F1-score, and our architecture significantly decreases the amount of learnable parameters. This improvement over DeepConvLSTM decreases training time by as much as 48%. Our results stand in contrast to the belief that one needs at least a 2-layered LSTM when dealing with sequential data. Based on our results we argue that said claim might not be applicable to sensor-based HAR. CCS CONCEPTS• Human-centered computing → Ubiquitous and mobile computing design and evaluation methods; • Computing methodologies → Neural networks.