Applying 1D sensor DenseNet to Sussex-Huawei locomotion-transportation recognition challenge

Zhu, Yida; Zhao, Fang; Chen, Runze

doi:10.1145/3341162.3345571

Cited by 13 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, GRU is also proven to be very efficient in mode detection problem [12]. Similar to LSTM and GRU, 1D convolutional layers have been demonstrated in handling time-series of mode detection, as demonstrated by 1D DenseNet proposed in [24].…”

Section: Deep Learning Methodsmentioning

confidence: 97%

“…More commonly, a dataloader module is called to directly prepare the input from raw data. This module includes functions to compute the magnitude of variables like acceleration, gyroscope, and magnetic field [24], jerk of the sensors [5], and adapting the mobile phone's coordinate system to a global reference frame, among others [8]. In this way, signal expert knowledge is not necessarily required to pre-process on the raw data, which makes it considerably distinct from the feature extraction concept discussed earlier.…”

Section: Deep Learning Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Deep CNN-BiLSTM Model for Transportation Mode Detection Using Smartphone Accelerometer and Magnetometer

Tang

Jahan

Roth

2022

2022 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Transportation mode detection from smartphone data is investigated as a relevant problem in the multi-modal transportation systems context. Neural networks are chosen as a timely and viable solution. The goal of this paper is to solve such a problem with a combination model of Convolutional Neural Network (CNN) and Bidirectional-Long short-term memory (BiLSTM) only processing accelerometer and magnetometer data. The performance in terms of accuracy and F1 score on the Sussex-Huawei Locomotion-Transportation (SHL) challenge 2018 dataset is comparable to methods that require the processing of a wider range of sensors. The uniqueness of our work is the light architecture requiring less computational resources for training and consequently a shorter inference time.

show abstract

Section: Deep Learning Methodsmentioning

confidence: 97%

Section: Deep Learning Methodsmentioning

confidence: 99%

Deep CNN-BiLSTM Model for Transportation Mode Detection Using Smartphone Accelerometer and Magnetometer

Tang

Jahan

Roth

2022

2022 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

show abstract

“…Charissa et al [29] employed a CNN using filters with a large time span to explore the long temporal correlation, and pooling over time is gradually used alternating with convolutional layers to reduce the loss over time. Zhu et al [16] proposed to use a 1D DenseNet model in order to take advantage of deeper CNNs. The DenseNet is first applied on each sensor independently and then combined together.…”

Section: Deep Learning Based Methodsmentioning

confidence: 99%

“…Especially with the rapid development of deep learning, lots of convolutional neural network (CNN) and recurrent neural network (RNN) based methods have been developed in the last few years. For the CNN based methods, EmbraceNet [15] and DenseNet [16] have been proposed for the task. However, due to the nature of convolution, its receptive field in the time domain is relatively small and the long-range temporal information cannot be well captured.…”

Section: Introductionmentioning

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

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

“…The user's HAR context expresses semantic information about real-time user activity from the original sensor signal sequence. User activity information is an essential context for many applications, such as smart homes [1], [2], human-computer interaction [3], health monitoring [4], [5], transportation schedules [6], [7], [8], etc. Those applications require real-time response and accurate HAR performance.…”

Section: Introductionmentioning

confidence: 99%

A Light-weight Deep Human Activity Recognition Algorithm Using Multi-knowledge Distillation

Chen¹,

Luo²,

Zhao³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Human activity recognition (HAR) based on IMU sensors is an essential domain in ubiquitous computing. Because of the improving trend to deploy artificial intelligence into IoT devices or smartphones, more researchers design the HAR models for embedded devices. We propose a plug-and-play HAR modeling pipeline with multi-level distillation to build deep convolutional HAR models with native support of embedded devices. SMLDist consists of stage distillation, memory distillation, and logits distillation, which covers all the information flow of the deep models. Stage distillation constrains the learning direction of the intermediate features.Memory distillation teaches the student models how to explain and store the inner relationship between high-dimensional features based on Hopfield networks. Logits distillation constructs distilled logits by a smoothed conditional rule to keep the probable distribution and improve the correctness of the soft target. We compare the performance of accuracy, F1 macro score, and energy cost on the embedded platform of various state-of-the-art HAR frameworks with a MobileNet V3 model built by SMLDist. The produced model has well balance with robustness, efficiency, and accuracy. SMLDist can also compress the models with minor performance loss in an equal compression rate than other state-of-the-art knowledge distillation methods on seven public datasets.

show abstract

Applying 1D sensor DenseNet to Sussex-Huawei locomotion-transportation recognition challenge

Cited by 13 publications

References 6 publications

Deep CNN-BiLSTM Model for Transportation Mode Detection Using Smartphone Accelerometer and Magnetometer

Deep CNN-BiLSTM Model for Transportation Mode Detection Using Smartphone Accelerometer and Magnetometer

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

A Light-weight Deep Human Activity Recognition Algorithm Using Multi-knowledge Distillation

Contact Info

Product

Resources

About