Facial expression recognition with FRR‐CNN

Xie, Siyue; Hu, Haifeng

doi:10.1049/el.2016.4328

Cited by 73 publications

(34 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Ten‐fold cross‐validation is used test the accuracy of the models . The input data are divided into ten equal portions, of which nine portions are taken as training data and one portion are taken as testing data in turn.…”

Section: Emotional Evaluation Modelsmentioning

confidence: 99%

“…Ten-fold cross-validation is used test the accuracy of the models. 59 The input data are divided into ten equal portions, of which nine portions are taken as training data and one portion are taken as testing data in turn. Root mean square error (RMSE) is selected to evaluate the accuracy of the emotional evaluation models, as shown in Equation (2).…”

Section: Model Validationmentioning

confidence: 99%

See 1 more Smart Citation

Research on color optimization of tricolor product considering color harmony and users' emotion

Guo

Nagamachi

et al. 2019

Color Research & Application

View full text Add to dashboard Cite

There are few studies on the tricolor design optimization, and the influence of color‐area ratio on users' emotion has been ignored. This article aims to achieve multiobjective optimization of tricolor product color design. Two modes of color‐area ratio are put forward. Using the proposed method of generating tricolor schemes, 368 tricolor schemes of the representative baby carriage are designed. Through questionnaire survey, color images are screened, and combined with correlation analysis and factor analysis, five perceptual features of color design are determined as “Order,” “Excitement,” “Temperature,” “Color harmony,” and “Users' emotional preference.” From two perspectives of color information processing, radial basis function neural networks are used to construct two emotional evaluation models. By integrating the radial basis function neural networks and a genetic algorithm, this study achieves multiobjective optimization of tricolor product color design for two optimization objectives of color harmony and users' emotional preference. Verification results show that the optimization schemes are significantly better than other schemes. Due to similar product characteristics, children's electric car is utilized to verify the generalization capability of the optimization method proposed in this article. It has been demonstrated that the optimization schemes achieve higher scores than randomly selected color schemes and the actual subjective scores match with the predicted scores computed by the color optimization method.

show abstract

Section: Emotional Evaluation Modelsmentioning

confidence: 99%

Section: Model Validationmentioning

confidence: 99%

Research on color optimization of tricolor product considering color harmony and users' emotion

Guo

Nagamachi

et al. 2019

Color Research & Application

View full text Add to dashboard Cite

show abstract

“…For instance, Kim et al [32] proposed a hierarchical committee of deep CNNs by combining the decisions of multiple models trained on public FER databases. A feature redundancy-reduced (FRR-CNN) is proposed by Xie et al [33] for FER to generate less redundant features and compact representation of the image. Uddin et al [34] extracted local directional position patterns from depth video data and fed them into a deep belief network (DBN) for FER.…”

Section: Introductionmentioning

confidence: 99%

Personalized Movie Summarization Using Deep CNN‐Assisted Facial Expression Recognition

et al. 2019

View full text Add to dashboard Cite

Personalized movie summarization is demand of the current era due to an exponential growth in movies production. The employed methods for movies summarization fail to satisfy the user's requirements due to the subjective nature of movies data. Therefore, in this paper, we present a user-preference based movie summarization scheme. First, we segmented movie into shots using a novel entropy-based shots segmentation mechanism. Next, temporal saliency of shots is computed, resulting in highly salient shots in which character faces are detected. The resultant shots are then forward propagated to our trained deep CNN model for facial expression recognition (FER) to analyze the emotional state of the characters. The final summary is generated based on userpreferred emotional moments from the seven emotions, i.e., afraid, angry, disgust, happy, neutral, sad, and surprise. The subjective evaluation over five Hollywood movies proves the effectiveness of our proposed scheme in terms of user satisfaction. Furthermore, the objective evaluation verifies the superiority of the proposed scheme over state-of-the-art movie summarization methods.

show abstract

“…With the recent application of deep learning, FER has made significant progress. Several works focus on the use of Convolutional Neural Networks (CNNs) for collective feature extraction and detection of prototypical emotions (Dachapally, 2017;Xie and Hu, 2017). CNNs constitute an end to end model which take an image input and perform combined feature extraction and classification within a single stage (Liu, Zhang and Pan, 2016).…”

Section: Introductionmentioning

confidence: 99%

Human Centric Facial Expression Recognition

Clawson

Delicato

Bowerman

2018

Electronic Workshops in Computing

View full text Add to dashboard Cite

Facial expression recognition (FER) is an area of active research, both in computer science and in behavioural science. Across these domains there is evidence to suggest that humans and machines find it easier to recognise certain emotions, for example happiness, in comparison to others. Recent behavioural studies have explored human perceptions of emotion further, by evaluating the relative contribution of features in the face when evaluating human sensitivity to emotion. It has been identified that certain facial regions have more salient features for certain expressions of emotion, especially when emotions are subtle in nature. For example, it is easier to detect fearful expressions when the eyes are expressive. Using this observation as a starting point for analysis, we similarly examine the effectiveness with which knowledge of facial feature saliency may be integrated into current approaches to automated FER. Specifically, we compare and evaluate the accuracy of 'full-face' versus upper and lower facial area convolutional neural network (CNN) modelling for emotion recognition in static images, and propose a human centric CNN hierarchy which uses regional image inputs to leverage current understanding of how humans recognise emotions across the face. Evaluations using the CK+ dataset demonstrate that our hierarchy can enhance classification accuracy in comparison to individual CNN architectures, achieving overall true positive classification in 93.3% of cases.

show abstract

Facial expression recognition with FRR‐CNN

Cited by 73 publications

References 10 publications

Research on color optimization of tricolor product considering color harmony and users' emotion

Research on color optimization of tricolor product considering color harmony and users' emotion

Personalized Movie Summarization Using Deep CNN‐Assisted Facial Expression Recognition

Human Centric Facial Expression Recognition

Contact Info

Product

Resources

About