Hybrid Facial Emotion Recognition Using CNN-Based Features

Shahzad, Hamza; Bhatti, Sohail Masood; Jaffar, Arfan; Akram, Sheeraz; Alhajlah, Mousa; Mahmood, Awais

doi:10.3390/app13095572

Cited by 11 publications

(3 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shahzad and colleagues [62] used Convolutional Neural Networks (CNN) for emotion classification. The simulation findings show that the AlexNet and VGG-16 architectures outperform the SoftMax classifier in terms of support vector machine (SVM) and ensemble classifier performance.…”

Section: B) Object Detectionmentioning

confidence: 99%

Privacy-Preserving Photo Sharing on Online Social Networks: A Review

Sajid,

Kavitha

2024

IJSSE

View full text Add to dashboard Cite

Online social networking applications and their services have become an essential part of the human lifestyle. These social services help online users in different ways, such as through social visibility, content sharing, communication, promotions, etc. In the same way, the services of online social networks (OSNs) pose privacy risks for their users. The service of photo sharing in OSNs, in particular, causes leakage of online social users' personal information. It's critical to comprehend previous studies that looked at privacypreserving photo sharing and whether users have their privacy protected when sharing photos. The goal of this review is to bridge the gap between the growing demand for image sharing via OSNs and individualized privacy requirements. This effort presents a comprehensive analysis of "privacy-preserving" technologies that specifically address contemporary privacy concerns associated with sharing images on online social networks (OSNs). This study presents a comprehensive analysis framework that focuses on the complete lifecycle of image sharing on online social networks (OSNs). This work presented a review framework called Privacy-Preserving Photo Sharing (PPPS) by categorizing previous works into three stages related to online secure photo sharing. Preprocessing, privacy settings, and photo publishing are the stages used in this survey's design to secure photo sharing. This framework aims to tackle the many privacy challenges and propose appropriate solutions in this multidisciplinary domain. During each phase, we analyse common user behaviours connected to sharing, the privacy concerns that arise from those behaviours, and evaluate representative solutions that are proposed by previous works.

show abstract

Section: B) Object Detectionmentioning

confidence: 99%

Privacy-Preserving Photo Sharing on Online Social Networks: A Review

Sajid,

Kavitha

2024

IJSSE

View full text Add to dashboard Cite

show abstract

“…With the continuous development of convolutional networks, the facial expression recognition models proposed based on this technique have achieved excellent performance [7][8][9]. However, the spatial localization of convolutional networks makes it difficult for the model to learn the dependencies between different facial regions, thus limiting the understanding of global facial expressions.…”

Section: Introductionmentioning

confidence: 99%

Context Transformer and Adaptive Method with Visual Transformer for Robust Facial Expression Recognition

Xiong,

Zhang,

Zheng

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

In real-world scenarios, the facial expression recognition task faces several challenges, including lighting variations, image noise, face occlusion, and other factors, which limit the performance of existing models in dealing with complex situations. To cope with these problems, we introduce the CoT module between the CNN and ViT frameworks, which improves the ability to perceive subtle differences by learning the correlations between local area features at a fine-grained level, helping to maintain the consistency between the local area features and the global expression, and making the model more adaptable to complex lighting conditions. Meanwhile, we adopt an adaptive learning method to effectively eliminate the interference of noise and occlusion by dynamically adjusting the parameters of the Transformer Encoder’s self-attention weight matrix. Experiments demonstrate the accuracy of our CoT_AdaViT model in the Oulu-CASIA dataset as (NIR: 87.94%, VL: strong: 89.47%, weak: 84.76%, dark: 82.28%). As well as, CK+, RAF-DB, and FERPlus datasets achieved 99.20%, 91.07%, and 90.57% recognition results, which achieved excellent performance and verified that the model has strong recognition accuracy and robustness in complex scenes.

show abstract

“…As artificial intelligence continues to advance rapidly, there has been a noticeable shift towards emphasizing the importance of emotional intelligence. Nowadays, more and more individuals emphasize developing and understanding their emotional intelligence, in tandem with technological progressions [1,2]. Facial expressions are incredibly informative as they provide a window into a person's emotions, character, and intentions.…”

Section: Introductionmentioning

confidence: 99%

Multi-Modal CNN Features Fusion for Emotion Recognition: A Modified Xception Model

Shahzad,

Bhatti,

Jaffar

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

Facial expression recognition (FER) is advancing human-computer interaction, especially, today, where facial masks are commonly worn due to the COVID-19 pandemic. Traditional unimodal techniques for facial expression recognition may be ineffective under these circumstances. To address this challenge, multimodal approaches that incorporate data from various modalities, such as voice expressions, have emerged as a promising solution. This paper proposed a novel multimodal methodology based on deep learning to recognize facial expressions under masked conditions effectively. The approach utilized two standard datasets, M-LFW-F and CREMA-D, to capture facial and vocal emotional expressions. A multimodal neural network was then trained using fusion techniques, outperforming conventional unimodal methods in facial expression recognition. Experimental evaluations demonstrated that the proposed approach achieved an accuracy of 79.81%, a significant improvement over the 68.81% accuracy attained by the unimodal technique. These results highlight the superior performance of the proposed approach in facial expression recognition under masked conditions.INDEX TERMS Deep learning, multimodal fusion techniques, neural network, facial expression under the mask.

show abstract

Hybrid Facial Emotion Recognition Using CNN-Based Features

Cited by 11 publications

References 47 publications

Privacy-Preserving Photo Sharing on Online Social Networks: A Review

Privacy-Preserving Photo Sharing on Online Social Networks: A Review

Context Transformer and Adaptive Method with Visual Transformer for Robust Facial Expression Recognition

Multi-Modal CNN Features Fusion for Emotion Recognition: A Modified Xception Model

Contact Info

Product

Resources

About