Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors

Li, Frédéric; Shirahama, Kimiaki; Nisar, Muhammad Adeel; Köping, Lukas; Grzegorzek, Marcin

doi:10.3390/s18020679

Cited by 246 publications

(224 citation statements)

References 31 publications

Supporting

Mentioning

219

Contrasting

Unclassified

Order By: Relevance

“…Through the above segment approach, each "picture" consists of a T×S matrix. We set the segment parameters like [44], use a time-window of 2s on the OPPORTUNITY and PAMAP2 datasets, resulting in T=64, and σ = 3. On the UniMiB-SHAR dataset, a time-window of 2s was used, resulting in T=96.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition

et al. 2018

View full text Add to dashboard Cite

Deep Convolutional Neural Networks (DCNNs) are currently popular in human activity recognition (HAR) applications. However, in the face of modern artificial intelligence sensor-based games, many research achievements cannot be practically applied on portable devices (i.e., smart phone, VR/AR). DCNNs are typically resource-intensive and too large to be deployed on portable devices, thus this limits the practical application of complex activity detection. In addition, since portable devices do not possess high-performance Graphic Processing Units (GPUs), there is hardly any improvement in Action Game (ACT) experience. Besides, in order to deal with multi-sensor collaboration, all previous human activity recognition models typically treated the representations from different sensor signal sources equally. However, distinct types of activities should adopt different fusion strategies. In this paper, a novel scheme is proposed. This scheme is used to train 2-bit Convolutional Neural Networks with weights and activations constrained to {-0.5, 0, 0.5}. It takes into account the correlation between different sensor signal sources and the activity types. This model, which we refer to as DFTerNet, aims at producing a more reliable inference and better trade-offs for practical applications. It's known that quantization of weights and activations can substantially reduce memory size and use more efficient bitwise operations to replace floating or matrix operations to achieve much faster calculation and lower power consumption. Our basic idea is to exploit quantization of weights and activations directly in pre-trained filter banks and adopt dynamic fusion strategies for different activity types. Experiments demonstrate that by using a dynamic fusion strategy, it is possible to exceed the baseline model performance by up to ∼5% on activity recognition datasets like the OPPORTUNITY and PAMAP2 datasets. Using the quantization method proposed, we were able to achieve performances closer to that of the full-precision counterpart. These results were also verified using the UniMiB-SHAR dataset. In addition, the proposed method can achieve ∼9× acceleration on CPUs and ∼11× memory saving.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Early fusion. All joints from multi-sensors in different parts are stacked as input of the network [22,44].…”

Section: Scalability To Multiple Sensors (Imus)mentioning

confidence: 99%

DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The learning-based methods include AE [24], MLP [25], CNN [14], LSTM [26], Hybrid [27], ResNet [20]. As in conventional methods, we use hand-crafted features, readers can find more details in [37]. For learning-based methods, we use raw activity data as input.…”

Section: Baselinementioning

confidence: 99%

“…For learning-based methods, we use raw activity data as input. Follow by [37], the hyper-parameters of these learning-based baseline models except ResNet 2 for the OPPORTUNITY and UniMiB-SHAR datasets are provided in Table 6. iii. Implementation and Setting…”

Section: Baselinementioning

confidence: 99%

Asymmetric Residual Neural Network for Accurate Human Activity Recognition

et al. 2019

View full text Add to dashboard Cite

Human Activity Recognition (HAR) using deep neural network has become a hot topic in human-computer interaction. Machine can effectively identify human naturalistic activities by learning from a large collection of sensor data. Activity recognition is not only an interesting research problem, but also has many real-world practical applications. Based on the success of residual networks in achieving a high level of aesthetic representation of the automatic learning, we propose a novel Asymmetric Residual Network, named ARN. ARN is implemented using two identical path frameworks consisting of (1) a short time window, which is used to capture spatial features, and (2) a long time window, which is used to capture fine temporal features. The long time window path can be made very lightweight by reducing its channel capacity, yet still being able to learn useful temporal representations for activity recognition. In this paper, we mainly focus on proposing a new model to improve the accuracy of HAR. In order to demonstrate the effectiveness of ARN model, we carried out extensive experiments on benchmark datasets (i.e., OPPORTUNITY, UniMiB-SHAR) and compared with some conventional and state-of-the-art learning-based methods. Then, we discuss the influence of networks parameters on performance to provide insights about its optimization. Results from our experiments show that ARN is effective in recognizing human activities via wearable datasets.

show abstract

“…The drawback of these methods is that they rely heavily on human experience or domain knowledge. In recent years, with the rapid development of deep learning technology, the classification performance of HAR based on deep learning networks has increased substantially [4,5]. Compared with traditional methods, deep learning networks can automatically extract high-dimensional features from raw sensor inputs.…”

Section: Introductionmentioning

confidence: 99%

Margin-Based Deep Learning Networks for Human Activity Recognition

Wang

Jin

et al. 2020

Sensors

View full text Add to dashboard Cite

Human activity recognition (HAR) is a popular and challenging research topic, driven by a variety of applications. More recently, with significant progress in the development of deep learning networks for classification tasks, many researchers have made use of such models to recognise human activities in a sensor-based manner, which have achieved good performance. However, sensor-based HAR still faces challenges; in particular, recognising similar activities that only have a different sequentiality and similarly classifying activities with large inter-personal variability. This means that some human activities have large intra-class scatter and small inter-class separation. To deal with this problem, we introduce a margin mechanism to enhance the discriminative power of deep learning networks. We modified four kinds of common neural networks with our margin mechanism to test the effectiveness of our proposed method. The experimental results demonstrate that the margin-based models outperform the unmodified models on the OPPORTUNITY, UniMiB-SHAR, and PAMAP2 datasets. We also extend our research to the problem of open-set human activity recognition and evaluate the proposed method’s performance in recognising new human activities.

show abstract

Comparison of Feature Learning Methods for Human Activity Recognition Using Wearable Sensors

Cited by 246 publications

References 31 publications

DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition

DFTerNet: Towards 2-bit Dynamic Fusion Networks for Accurate Human Activity Recognition

Asymmetric Residual Neural Network for Accurate Human Activity Recognition

Margin-Based Deep Learning Networks for Human Activity Recognition

Contact Info

Product

Resources

About