Virtual facial expression recognition using deep CNN with ensemble learning

Chirra, Venkata Rami Reddy; Reddy, U. Srinivasulu; Kolli, Venkata Krishna Kishore

doi:10.1007/s12652-020-02866-3

Cited by 37 publications

(11 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The group ensemble [12] approach in which all ensemble models are interconnected to a single ConvNet instead trained separately in multiple ConvNets to reduce the computation cost. Four block multi-block deep convolutional neural networks (CNN) [13] use two pretrained VGG16 and ResNet models which are utilized separately for extracting the features of images and these features are ensemble through three models like SVM, random forest, and linear regression to improve recognition accuracy performance. Image classification using Multi-Layers Deep Features Fusion and selection-based technique contributed significantly by fusion of VGG and Inception V3 deep learning feature set and selecting a sturdy feature set using Multi Logistic Regression controlled Entropy-Variances method [14].…”

Section: Related Workmentioning

confidence: 99%

Enhancing Marathi Handwritten Character Recognition Using Ensemble Learning

Chikmurge¹,

Raghunathan²

2023

View full text Add to dashboard Cite

In the current era, the Optical Character Recognition (OCR) model plays a vital role in converting images of handwritten characters or words into text editable script. During the COVID-19 pandemic, students' performance is assessed based on multiple-choice questions and handwritten answers so, in this situation, the need for handwritten recognition has become acute. Handwritten answers in any regional language need the OCR model to transform the readable machine-encoded text for automatic assessment which will reduce the burden of manual assessment. The single Convolutional Neural Network (CNN) algorithm recognizes the handwritten characters but its accuracy is suppressed when dataset volume is increased. In proposed work stacking and soft voting ensemble mechanisms that address multiple CNN models to recognize the handwritten characters. The performance of the ensemble mechanism is significantly better than the single CNN model. This proposed work ensemble VGG16, Alexnet and LeNet-5 as base classifiers using stacking and soft voting ensemble approaches. The overall accuracy of the proposed work is 98.66% when the soft voting ensemble has three CNN classifiers.

show abstract

Section: Related Workmentioning

confidence: 99%

Enhancing Marathi Handwritten Character Recognition Using Ensemble Learning

Chikmurge¹,

Raghunathan²

2023

View full text Add to dashboard Cite

show abstract

“…Synthetic datasets have shown to be a good replacement for real-image datasets since they achieve recognition rates that are comparable to the genuine ones [ 13 , 14 ]. In particular, the UIBVFED dataset has been used in several FER and Emotion Recognition (ER) studies [ 15 , 16 ]. However, despite the large number of facial expressions categorized, this original version of the dataset has a limited number of characters, i.e., a limited number of samples per category.…”

Section: Related Workmentioning

confidence: 99%

UIBVFEDPlus-Light: Virtual facial expression dataset with lighting

Mascaró-Oliver,

Amengual-Alcover,

Roig-Maimó

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

It is well-known that lighting conditions have an important influence on the automatic recognition of human expressions. Although the impact of lighting on the perception of emotions has been studied in different works, databases of facial expressions do not consider intentional lighting. In this work, a new database of facial expressions performed by virtual characters with four different lighting configurations is presented. This database, named UIBVFEDPlus-Light, is an extension of the previously published UIBVFED virtual facial expression dataset. It includes 100 characters, four lighting configurations and a software application that allows one to interactively visualize the expressions, and manage their intensity and lighting condition. Also, an experience of use is described to show how this work can raise new challenges to facial expression and emotion recognition techniques under usual lighting environments. Thus, opening new study perspectives in this area.

show abstract

“…The texture features here were extracted by means of Gray-level Cooccurrence matrix therefore ensuring accuracy and robustness in overall classification. In [7], multi-block deep convolutional neural network was designed for facial expression recognition.…”

Section: Related Workmentioning

confidence: 99%

“…𝑅𝑆𝑆(𝛼 𝑘 |𝛼 0 ,𝛼 1 ,…,𝛼 𝑛 ) 𝑀𝑆𝐸 (7) From the above equation ( 7), if feature variable '𝑃𝐹𝐼 𝑗 ' is added to the model then the feature selection procedure determines whether the feature variable '𝑃𝐹𝐼 𝑖 ' should be eliminated or retained. This is evaluated by estimating the Aggregate F-statistic is mathematically stated as given below.…”

Section: 𝑓 𝑘 =mentioning

confidence: 99%