Ensemble learning for human activity recognition

Sekiguchi, Ryoichi; Abe, Kenji; Yokoyama, Takumi; Kumano, Masayasu; Kawakatsu, Masaki

doi:10.1145/3410530.3414346

Cited by 19 publications

(10 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method is further compared with the existing methods [39][40][41][42][43][44][45][46][47][48][49][50]. The results are shown in Table 4 including comparisons with the existing machine learning and deep learning methods.…”

Section: Comparison With State-of-the-art Classification Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A Framework of Combining Short-Term Spatial/Frequency Feature Extraction and Long-Term IndRNN for Activity Recognition

Zhao¹,

Li²,

Gao³

et al. 2020

Preprint

View full text Add to dashboard Cite

Smartphone sensors based human activity recognition is attracting increasing interests nowadays with the popularization of smartphones. With the high sampling rates of smartphone sensors, it is a highly long-range temporal recognition problem, especially with the large intra-class distances such as the smartphones carried at different locations such as in the bag or on the body, and the small inter-class distances such as taking train or subway. To address this problem, we propose a new framework of combining short-term spatial/frequency feature extraction and a long-term Independently Recurrent Neural Network (IndRNN) for activity recognition. Considering the periodic characteristics of the sensor data, short-term temporal features are first extracted in the spatial and frequency domains. Then the IndRNN, which is able to capture long-term patterns, is used to further obtain the long-term features for classification. In view of the large differences when the smartphone is carried at different locations, a group based location recognition is first developed to pinpoint the location of the smartphone. The Sussex-Huawei Locomotion (SHL) dataset from the SHL Challenge is used for evaluation. An earlier version of the proposed method has won the second place award in the SHL Challenge 2020 (the first place if not considering multiple models fusion approach). The proposed method is further improved in this paper and achieves 80.72% accuracy, better than the existing methods using a single model.

show abstract

Section: Comparison With State-of-the-art Classification Methodsmentioning

confidence: 99%

“…GAN [50] 34.4% Multi-View CNN [49] 37.3% Logistic regression [46] 55.7% InceptionTime [42] 69.4% 3-Layer CNN [41] 76.4% CNN + LSTM [48] 52.8% DenseNetX + GRU (Model Fusion based) [39] 88.5%…”

Section: Methods Performancementioning

confidence: 99%

A Framework of Combining Short-Term Spatial/Frequency Feature Extraction and Long-Term IndRNN for Activity Recognition

Zhao¹,

Li²,

Gao³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Several submissions converted the sensor data from phonecentred coordinate system to human-centred coordination system, which can potentially increase the robustness to phone placement (Zhu et al, 2020;Zhao et al, 2020;Siraj , 2020;Tseng et al, 2020;Sekiguchi et al, 2020;Ahmed et al, 2019). The idea with this technique is to reduce the dependence on the orientation of the phone relative to the body.…”

Section: Coordinate Transformationmentioning

confidence: 99%

“…To exploit this fact, several submissions employed machine learning techniques to recognize the phone location first, and then developed a position-dependent model using the training and validation data. Interestingly, most of these submissions can estimate the phone location ("Hips") correctly (Kalabakov et al, 2020;Widhalm et al, 2020;Zhao et al, 2020;Yaguchi et al, 2020;Sekiguchi et al, 2020), whereas one submission obtained a wrong estimation of the phone location ("Hand") (Siraj and et al, 2020).…”

Section: Position-specific Modelingmentioning

confidence: 99%

Three-Year Review of the 2018–2020 SHL Challenge on Transportation and Locomotion Mode Recognition From Mobile Sensors

Wang

Gjoreski

Ciliberto

et al. 2021

Front. Comput. Sci.

View full text Add to dashboard Cite

The Sussex-Huawei Locomotion-Transportation (SHL) Recognition Challenges aim to advance and capture the state-of-the-art in locomotion and transportation mode recognition from smartphone motion (inertial) sensors. The goal of this series of machine learning and data science challenges was to recognize eight locomotion and transportation activities (Still, Walk, Run, Bus, Car, Train, Subway). The three challenges focused on time-independent (SHL 2018), position-independent (SHL 2019) and user-independent (SHL 2020) evaluations, respectively. Overall, we received 48 submissions (out of 93 teams who registered interest) involving 201 scientists over the three years. The survey captures the state-of-the-art through a meta-analysis of the contributions to the three challenges, including approaches, recognition performance, computational requirements, software tools and frameworks used. It was shown that state-of-the-art methods can distinguish with relative ease most modes of transportation, although the differentiating between subtly distinct activities, such as rail transport (Train and Subway) and road transport (Bus and Car) still remains challenging. We summarize insightful methods from participants that could be employed to address practical challenges of transportation mode recognition, for instance, to tackle over-fitting, to employ robust representations, to exploit data augmentation, and to exploit smart post-processing techniques to improve performance. Finally, we present baseline results to compare the three challenges with a unified recognition pipeline and decision window length.

show abstract

“…Our target is developing the movement state classification algorithm using the SHL dataset 2021 [1]. We participated in the SHL recognition challenge with the team TDU_BSA in 2020 [2]. Accelerometers, geomagnetic sensors, gyro sensors, and barometric pressure sensors cannot be used in this challenge.…”

Section: Introductionmentioning

confidence: 99%

Phased Human Activity Recognition based on GPS

Sekiguchi

Abe

shogo

et al. 2021

Adjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of The

Self Cite

View full text Add to dashboard Cite

This paper describes an activity recognition method for Sussex-Huawei Locomotion-Transportation (SHL) recognition challenge by team TDU_BSA_BCI. The classification accuracy has been improved by switching the estimation model, depending on whether the location is available. Data, including location were classified by Deep Neural Network including LSTM layer. Data that exclude location were classified by the Gradient Boosting Decision Tree. The 2 outputs have been combined. They were optimized by applying a median filter. In the submission phase, the best F-measure obtained for the SHL validation-set was 65%. CCS CONCEPTS• Computing methodologies → Activity recognition and understanding.

show abstract

Ensemble learning for human activity recognition

Cited by 19 publications

References 6 publications

A Framework of Combining Short-Term Spatial/Frequency Feature Extraction and Long-Term IndRNN for Activity Recognition

A Framework of Combining Short-Term Spatial/Frequency Feature Extraction and Long-Term IndRNN for Activity Recognition

Three-Year Review of the 2018–2020 SHL Challenge on Transportation and Locomotion Mode Recognition From Mobile Sensors

Phased Human Activity Recognition based on GPS

Contact Info

Product

Resources

About