PSNet: Parametric Sigmoid Norm Based CNN for Face Recognition

Srivastava, Yash; Murali, Vaishnav; Dubey, Shiv Ram

doi:10.1109/cict48419.2019.9066169

Cited by 11 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For DA-CG-LSTM, there are three parameters, including window size T , batch size of input sequence, and the number of units m in the hidden layer of CG-LSTM. Set T ∈ [5,10,15,20,30,50], batch_size ∈ [16,32,64,128,256,512] and m ∈ [10,20,30,40,50]. Find the optimal parameters which are T = 15, batch_size = 128 and m = 30.…”

Section: B Parameter Setting and Evaluation Indicatorsmentioning

confidence: 99%

“…In order to effectively capture short-term mutation information, Parametric Sigmoid Norm Based CNN(PSNET) [20] was proposed to separate centroid of samples of different complexity by translating and scaling Sigmoid function, in order to extract relevant input sequences. However, it cannot solve the problem of Sigmoid function's oversaturation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Dual-Staged Attention Based Conversion-Gated Long Short Term Memory for Multivariable Time Series Prediction

Shi

Feng

2022

IEEE Access

View full text Add to dashboard Cite

In multivariate time series modeling, it is necessary to capture short-term mutation and longterm dependence information simultaneously. However, mechanism which can capture short-term change is difficult to be used to grasp long-term dependence information, and vice versa. In order to capture both short-term mutation and long-term dependence information in the same model, this paper proposed a dual-staged attention mechanism based on conversion-gated Long Short Term Memory network(DA-CG-LSTM). Hyperbolic tangent function is introduced into the input-gate and the forget-gate of Long Short Term Memory network(LSTM), which improves the ability of the network to extract the short-term mutation information. Further, dual-staged attention mechanism is added in the network, which includes input attention and temporal attention. Input attention adaptively extracts the feature relations of exogenous sequences, and temporal attention selects the relevant hidden layer states across all the time steps. Experiments on air quality and traffic flow time series data show that the proposed network has lower average absolute error, average absolute percentage error and root mean square error by more than 50% compared with Dual-staged Attention Recurrent Neural Network(DA-RNN) and Transformation-gated LSTM(TG-LSTM).INDEX TERMS Conversion-gated, dual-staged attention mechanism, long short-term memory network, time series prediction.

show abstract

Section: B Parameter Setting and Evaluation Indicatorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Dual-Staged Attention Based Conversion-Gated Long Short Term Memory for Multivariable Time Series Prediction

Shi

Feng

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To expand the unsaturated region of sigmoid activation function, we use the parametric sigmoid which allows some flexibility in network training in this paper. The parametric sigmoid function is defined as (6) [23]:…”

Section: Autoencoder and Classifiermentioning

confidence: 99%

Encoding Spectral-Spatial Features for Hyperspectral Image Classification in the Satellite Internet of Things System

Han

Chen

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Hyperspectral image classification is essential for satellite Internet of Things(IoT) to build a large scale land-cover surveillance system. After acquiring real-time land-cover information, the edge of the network transmits all the hyperspectral images by satellites with low-latency and high-efficiency to the cloud computing center, which are provided by satellite IoT. A gigantic amount of remote sensing data bring challenges to the storage and processing capacity of traditional satellite systems. When hyperspectral images are used in annotation of land-cover application, data dimension reduction for classifier efficiency often leads to the decrease of classifier accuracy, especially the region to be annotated consists of natural landform and artificial structure. This paper proposes encoding spectral-spatial features for hyperspectral image classification in the satellite Internet of Things system to extract features effectively, namely attribute profile stacked autoencoder (AP-SAE). Firstly, extended morphology attribute profiles EMAP is used to obtain spatial features of different attribute scales. Secondly, AP-SAE is used to extract spectral features with similar spatial attributes. In this stage the program can learn feature mappings, on which the pixels from the same land-cover class are mapped as closely as possible and the pixels from different land-cover categories are separated by a large margin. Finally, the program trains an effective classifier by using the network of the AP-SAE. Experimental results on three widely-used hyperspectral image (HSI) datasets and comprehensive comparisons with existing methods demonstrate that our proposed method can be used effectively in hyperspectral image classification.

show abstract

“…Recognition accuracy of test and validation data on FER2013 dataset. 5 ( †, ‡ and * denote the results reported inSrivastava et al [2019],Cao et al [2020],Xu et al [2021], respectively). The performance of GSC loss embedded into Arcface and Cosface is superior to all other methods in LFW, CALFW and CFP-FF/FP, indicating the proposed GSC loss is a benefit for achieving a SOTA result for the face verification tasks.…”

mentioning

confidence: 99%

GSC Loss: A Gaussian Score Calibrating Loss for Deep Learning

Zhao¹,

Dou²,

Ji³

et al. 2022

Preprint

View full text Add to dashboard Cite

Cross entropy (CE) loss integrated with softmax is an orthodox component in most classificationbased frameworks, but it fails to obtain an accurate probability distribution of predicted scores that is critical for further decision-making of poor-classified samples. The prediction score calibration provides a solution to learn the distribution of predicted scores which can explicitly make the model obtain a discriminative representation. Considering the entropy function can be utilized to measure the uncertainty of predicted scores. But, the gradient variation of it is not in line with the expectations of model optimization. To this end, we proposed a general Gaussian Score Calibrating (GSC) loss to calibrate the predicted scores produced by the deep neural networks (DNN). Extensive experiments on over 10 benchmark datasets demonstrate that the proposed GSC loss can yield consistent and significant performance boosts in a variety of visual tasks. Notably, our label-independent GSC loss can be embedded into common improved methods based on the CE loss easily.

show abstract

PSNet: Parametric Sigmoid Norm Based CNN for Face Recognition

Cited by 11 publications

References 24 publications

A Dual-Staged Attention Based Conversion-Gated Long Short Term Memory for Multivariable Time Series Prediction

A Dual-Staged Attention Based Conversion-Gated Long Short Term Memory for Multivariable Time Series Prediction

Encoding Spectral-Spatial Features for Hyperspectral Image Classification in the Satellite Internet of Things System

GSC Loss: A Gaussian Score Calibrating Loss for Deep Learning

Contact Info

Product

Resources

About