Feature Extraction and Pattern Recognition for Human Motion by a Deep Sparse Autoencoder

Liu, Hailong; Taniguchi, Tadahiro

doi:10.1109/cit.2014.144

Cited by 22 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, some studies are using ML models [27][28][29] and DL in human action recognition [5,30,31]. However, these studies only focus on developing RNN and Long Short Term Memory (LSTM) models to predict but do not care about the characteristics of the object and extraction feature methods …”

Section: Evaluation Of Resultsmentioning

confidence: 99%

“…Deep Learning (DL) is a new research trend in recent years for many applications, such as image processing, object detection, and remote control [1][2][3][4]. DL has two main models: Convolutional Neural Network (CNN) used to feature extraction in image processing [5,6], and Recurrent Neural Network (RNN) used to handle sequence identification (sequence/time-series) [7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Human action recognition method based on conformal geometric algebra and recurrent neural network

Nguyen

Pham

Hao

et al. 2020

ICS

View full text Add to dashboard Cite

Deep Learning (DL) plays an important role in machine learning and artificial intelligence. DL is widely applied in many fields with high dimensional data, including natural language processing, image recognition. High dimensional data can lead to problems in machine learning such as overfitting, degradation of accuracy. To address these issues, some methods, such as Principal Components Analysis (PCA), principal component regression (PCR), Multi-class Linear Discriminant Analysis (MLDA), were proposed to reduce dimensions of the data and computational complexity simultaneously. The drawback of these methods is that they only work well on data distributed on the plane. In the case of the data distributed on the hyper-sphere, such as objects moving in space, the processing results are not so good as expected. In this paper, we propose the use of Conformal Geometric Algebra (CGA) to extract features and simultaneously reduce the dimensionality of a dataset for human activity recognition using Recurrent Neural Network (RNN). First, human activity data in a 3-dimensional coordinate system is pre-processed and normalized by calculating deviations from the mean coordinate. Next, the data is transformed to vectors in CGA space and its dimensions are reduced to return the feature vectors. Finally, we use the RNN model to train feature vectors. Empirical results performed on the CMU eight actions dataset show that the CGA combined with RNN gave the best test results of 92.5%.

show abstract

Section: Evaluation Of Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Human action recognition method based on conformal geometric algebra and recurrent neural network

Nguyen

Pham

Hao

et al. 2020

ICS

View full text Add to dashboard Cite

show abstract

“…Therefore, it is easy to train an algorithm that can perform significantly on a specific input. In 2014, the study by Liu, & Taniguchi [118], applied deep sparse autoencoder that extracted low dimensional features, which represent the characteristics of individual motion efficiently, from data of human action or motion with high dimensional. Recently, the SAE ideal has become popular for generative models learning of data [119].…”

Section: B Sae For Pattern Recognitionmentioning

confidence: 99%

Comprehensive Review of Artificial Neural Network Applications to Pattern Recognition

et al. 2019

View full text Add to dashboard Cite

The era of artificial neural network (ANN) began with a simplified application in many fields and remarkable success in pattern recognition (PR) even in manufacturing industries. Although significant progress achieved and surveyed in addressing ANN application to PR challenges, nevertheless, some problems are yet to be resolved like whimsical orientation (the unknown path that cannot be accurately calculated due to its directional position). Other problem includes; object classification, location, scaling, neurons behavior analysis in hidden layers, rule, and template matching. Also, the lack of extant literature on the issues associated with ANN application to PR seems to slow down research focus and progress in the field. Hence, there is a need for state-of-the-art in neural networks application to PR to urgently address the abovehighlights problems for more successes. The study furnishes readers with a clearer understanding of the current, and new trend in ANN models that effectively addresses PR challenges to enable research focus and topics. Similarly, the comprehensive review reveals the diverse areas of the success of ANN models and their application to PR. In evaluating the performance of ANN models, some statistical indicators for measuring the performance of the ANN model in many studies were adopted. Such as the use of mean absolute percentage error (MAPE), mean absolute error (MAE), root mean squared error (RMSE), and variance of absolute percentage error (VAPE). The result shows that the current ANN models such as GAN, SAE, DBN, RBM, RNN, RBFN, PNN, CNN, SLP, MLP, MLNN, Reservoir computing, and Transformer models are performing excellently in their application to PR tasks. Therefore, the study recommends the research focus on current models and the development of new models concurrently for more successes in the field.INDEX TERMS Artificial neural networks, application to pattern recognition, feedforward neural networks, feedback neural networks, hybrid models.

show abstract

“…So deep SAE was adopted to reduce the dimension of text, by extracting the structure hidden in data. In Ref [20], the text feature was extracted by using PCA, a shallow SAE and a deep SAE. The result indicated that the deep SAE has higher recognition accuracy, better generalization and more stability.…”

Section: Representation Of Textual Information In Htpnnmentioning

confidence: 99%

Stock Volatility Prediction by Hybrid Neural Network

et al. 2019

View full text Add to dashboard Cite

Stock price volatility forecasting is a hot topic in time series prediction research, which plays an important role in reducing investment risk. However, the trend of stock price not only depends on its historical trend, but also on its related social factors. This paper proposes a hybrid time-series predictive neural network (HTPNN) that combines the effection of news. The features of news headlines are expressed as distributed word vectors which are dimensionally reduced to optimize the efficiency of the model by sparse automatic encoders. Then, according to the timeliness of stocks, the daily K-line data is combined with the news. HTPNN captures the potential law of stock price fluctuation by learning the fusion feature of news and time series, which not only retains the effective information of news and stock data, but also eliminates the redundant information of the text. Compared with the state-of-the-art methods, our method combines more abundant stock characteristics and has more advantages in running speed. Besides, the accuracy is averagely improved by nearly 5%. INDEX TERMS Stock prediction, news, natural language processing, hybrid neural network.

show abstract

Feature Extraction and Pattern Recognition for Human Motion by a Deep Sparse Autoencoder

Cited by 22 publications

References 10 publications

Human action recognition method based on conformal geometric algebra and recurrent neural network

Human action recognition method based on conformal geometric algebra and recurrent neural network

Comprehensive Review of Artificial Neural Network Applications to Pattern Recognition

Stock Volatility Prediction by Hybrid Neural Network

Contact Info

Product

Resources

About