Facial Recognition via Transfer Learning: Fine-Tuning Keras_vggface

Luttrell, Joseph; Zhou, Zhaoxian; Zhang, Chaoyang; Gong, Ping; Zhang, Yuanyuan

doi:10.1109/csci.2017.98

Cited by 13 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was developed using the VGGFace dataset. With 2.6 mil-lion face photos, VGGFace [21] is huge. We have frozen the entire network except for the batch normalization layers.…”

Section: 𝐷 = #(𝑦mentioning

confidence: 99%

Improved Methods for Automatic Facial Expression Recognition

Echoukairi¹,

Ghmary²,

Ziani³

et al. 2023

Int. J. Interact. Mob. Technol.

View full text Add to dashboard Cite

Facial expressions constitute one of the most effective and instinctive methods that allow people to communicate their emotions and intentions. In this context, the both Machine Learning (ML) and Convolutional Neural Networks (CNNs) have been used for emotion recognition. Efficient recognition systems are required for good human-computer interaction. However, facial expression recognition is related to several methods that impact the performance of facial recognition systems. In this paper, we demonstrate a state-of-the-art of 65% accuracy on the FER2013 dataset by leveraging numerous techniques from recent research and we also proposed some new methods for improving accuracy by combining CNN architectures such as VGG-16 and Resnet-50 with auxiliary datasets such as JAFFE and CK. To predict emotions, we used a second approach based on geometric features and facial landmarks to calculate and transmit the feature vector to the SVM model. The results show that the ResNet50 model outperforms all other emotion prediction models in real time by maximizing.

show abstract

“…It was developed using the VGGFace dataset. With 2.6 mil-lion face photos, VGGFace [21] is huge. We have frozen the entire network except for the batch normalization layers.…”

Section: 𝐷 = #(𝑦mentioning

confidence: 99%

Improved Methods for Automatic Facial Expression Recognition

Echoukairi¹,

Ghmary²,

Ziani³

et al. 2023

Int. J. Interact. Mob. Technol.

View full text Add to dashboard Cite

show abstract

“…The results are shown in Table 11. al, 2022;Taherkhani et al, 2019;Guo et al, 2017;Qin et al, 2019;Filippidou and Papakostas, 2020;Perti et al, 2020;Sepas-Moghaddam et al, 2020;Zhiqi, 2021;Tran et al, 2022;Luttrell et al, 2017;Nam et al, 2018;Chandran et al, 2018;Choi and Lee, 2020;Zangeneh et al, 2020;Kim et al, 2020) ResNet (He et al, 2016) (Ling et al, 2020;Li et al, 2022;Horng et al, 2022;Li et al, 2022;Kim et al, 2020;Feng et al, 2020;Arafah et al, 2020;Almabdy and Elrefaei, 2019;Setio Aji et al, 2022;Gruber et al, 2017;Hou et al, 2020;Filippidou and Papakostas, 2020;Alkanhal et al, 2023;Zhou et al, 2020) AlexNet (Krizhevsky et al, 2012) (Bukovčiková et al, 2017;Bussey et al, 2017;Liu et al, 2017;Khan et al, 2019b;Han, 2021;H et al, 2023;Almabdy and Elrefaei, 2019;Ma et al, 2018;Alhan...…”

Section: Assessment Of Q3: What Type Of Cnn Model Is Most Commonly Us...mentioning

confidence: 99%

“…(Zhang and Kakadiaris, 2017) CASIA-WebFace 84.60%-91.40% (Zhang and Kakadiaris, 2017) UHDB31 68.80%-79.30% (Luttrell et al, 2017) FERET 97.00% (Ara et al, 2017) Private 95.00% (Hu et al, 2018) LFW 85.05%-95.77% (Yu et al, 2018) LFW 95.56%-99.01% (Chandran et al, 2018) Private 99.41% (Ding and Tao, 2018) CASIA-WebFace 94.96%-99.33% (Nam et al, 2018) LFW 98.87% (Bong et al, 2018) LFW 97.00% (Yang et al, 2018) Private 98.98% (Zhou et al, 2018) Private 94.10%-97.00% (Senthilkumar et al, 2018) UWA-HSFD 97.30% (Phankokkruad, 2018) Private 99.58% (Gilani and Mian, 2018) Private 98.74% (Zeng et al, 2018b) CNBC 98.45% (Zeng et al, 2018b) FERET 94.59% (Zhang et al, 2018) AR 98.80% (Zhang et al, 2018) LFW 94.60%-94.80% (Rao et al, 2018) CFP-FF 99.43%-99.46% (Rao et al, 2018) LFW 97.26%-97.31% (Rao et al, 2018) CFP-FP 91.71%-93.39% (Ma et al, 2018) ORL GoogleNet 93.69% (Ma et al, 2018) ORL Alexnet 92.74% (Ma et al, 2018) ORL ZF-5net 87.68% (Ma et al, 2018) AR face GoogleNet 76.17% (Ma et al, 2018) AR face ZF-5net 72.52% (Ma et al, 2018) AR face Alexnet 69.44% (Zeng et al, 2018a) AR face 99.30%-100.00% (Zeng et al, 2018a) FERET 93.90% (Zeng et al, 2018a) Extend (Zangeneh et al, 2020) FERET 81.40%-96.70% (Zangeneh et al, 2020) LFW 76.30% (Zhou et al, 2020) CFP-FP 93.26% (Zhou et al, 2020) CALFW 90.17% (Zhou et al, 2020) CPLFW 82.08% (Son et al, 2020) Private 91.30%…”

Section: 60%mentioning

confidence: 99%

Convolutional Neural Networks for Face Recognition: A Systematic Literature Review

Barcic¹,

Grd²,

Tomičić³

2023

Preprint

View full text Add to dashboard Cite

This paper presents a comprehensive overview of Convolutional Neural Networks (CNNs) in the context of face recognition. By analyzing 150 research papers, we investigate major publication channels, temporal trends, prevalent CNN models, primary data sets, accuracy levels, primary research focuses, and future prospects for improving CNN-based facial recognition. A major focus is placed on identifying prevalent CNN architectures, techniques used for facial recognition and shedding light on the evolving landscape of CNN designs. Furthermore, we examine the datasets used for training and testing CNNs, and evaluate the accuracy levels achieved by these models. Lastly, we discuss future directions for enhancing CNN-based facial recognition, including addressing bias and fairness, improving robustness to environmental variations, privacy preservation, and exploring transfer learning and multimodal fusion. This paper serves as a valuable resource, summarizing major trends in CNN-based face recognition. It provides insights for researchers and practitioners, guiding future advancements in this rapidly evolving field.

show abstract

“…DeepID proposed a lightweight convolutional neural network (CNN)-based face recognizer, using an input resolution with smaller pixels than DeepFace [19]. VGGFace, which appeared later, learned a deep network structure consisting of 15 convolution layers using a data set for high-capacity face recognition made by itself through an Internet search [20]. In addition, various studies were conducted to improve the performance of face recognition models such as DeepID2, DeepID3, and GoogLeNet [21][22][23].…”

Section: Introductionmentioning

confidence: 99%

Driver Facial Expression Analysis Using LFA-CRNN-Based Feature Extraction for Health-Risk Decisions

et al. 2020

View full text Add to dashboard Cite

As people communicate with each other, they use gestures and facial expressions as a means to convey and understand emotional state. Non-verbal means of communication are essential to understanding, based on external clues to a person’s emotional state. Recently, active studies have been conducted on the lifecare service of analyzing users’ facial expressions. Yet, rather than a service necessary for everyday life, the service is currently provided only for health care centers or certain medical institutions. It is necessary to conduct studies to prevent accidents that suddenly occur in everyday life and to cope with emergencies. Thus, we propose facial expression analysis using line-segment feature analysis-convolutional recurrent neural network (LFA-CRNN) feature extraction for health-risk assessments of drivers. The purpose of such an analysis is to manage and monitor patients with chronic diseases who are rapidly increasing in number. To prevent automobile accidents and to respond to emergency situations due to acute diseases, we propose a service that monitors a driver’s facial expressions to assess health risks and alert the driver to risk-related matters while driving. To identify health risks, deep learning technology is used to recognize expressions of pain and to determine if a person is in pain while driving. Since the amount of input-image data is large, analyzing facial expressions accurately is difficult for a process with limited resources while providing the service on a real-time basis. Accordingly, a line-segment feature analysis algorithm is proposed to reduce the amount of data, and the LFA-CRNN model was designed for this purpose. Through this model, the severity of a driver’s pain is classified into one of nine types. The LFA-CRNN model consists of one convolution layer that is reshaped and delivered into two bidirectional gated recurrent unit layers. Finally, biometric data are classified through softmax. In addition, to evaluate the performance of LFA-CRNN, the performance was compared through the CRNN and AlexNet Models based on the University of Northern British Columbia and McMaster University (UNBC-McMaster) database.

show abstract

Facial Recognition via Transfer Learning: Fine-Tuning Keras_vggface

Cited by 13 publications

References 10 publications

Improved Methods for Automatic Facial Expression Recognition

Improved Methods for Automatic Facial Expression Recognition

Convolutional Neural Networks for Face Recognition: A Systematic Literature Review

Driver Facial Expression Analysis Using LFA-CRNN-Based Feature Extraction for Health-Risk Decisions

Contact Info

Product

Resources

About