Face recognition in unconstrained environment with CNN

Fredj, Hana Ben; Bouguezzi, Safa; Souani, Chokri

doi:10.1007/s00371-020-01794-9

Cited by 83 publications

(38 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Ben Fredj et al [129] (2020) used aggressive data augmentation with randomly perturbing information and complicated facial appearance conditions. One of the main ideas was to use the adaptive fusion strategy of softmax loss and center loss, improving performance, and making the model more flexible and efficient.…”

Section: Investigations Based On Googlenet Architecturementioning

confidence: 99%

Past, Present, and Future of Face Recognition: A Review

et al. 2020

View full text Add to dashboard Cite

Face recognition is one of the most active research fields of computer vision and pattern recognition, with many practical and commercial applications including identification, access control, forensics, and human-computer interactions. However, identifying a face in a crowd raises serious questions about individual freedoms and poses ethical issues. Significant methods, algorithms, approaches, and databases have been proposed over recent years to study constrained and unconstrained face recognition. 2D approaches reached some degree of maturity and reported very high rates of recognition. This performance is achieved in controlled environments where the acquisition parameters are controlled, such as lighting, angle of view, and distance between the camera–subject. However, if the ambient conditions (e.g., lighting) or the facial appearance (e.g., pose or facial expression) change, this performance will degrade dramatically. 3D approaches were proposed as an alternative solution to the problems mentioned above. The advantage of 3D data lies in its invariance to pose and lighting conditions, which has enhanced recognition systems efficiency. 3D data, however, is somewhat sensitive to changes in facial expressions. This review presents the history of face recognition technology, the current state-of-the-art methodologies, and future directions. We specifically concentrate on the most recent databases, 2D and 3D face recognition methods. Besides, we pay particular attention to deep learning approach as it presents the actuality in this field. Open issues are examined and potential directions for research in facial recognition are proposed in order to provide the reader with a point of reference for topics that deserve consideration.

show abstract

Section: Investigations Based On Googlenet Architecturementioning

confidence: 99%

Past, Present, and Future of Face Recognition: A Review

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Over the last few years the face recognition accuracy has drastically improved. Different FR methods have been evolved over time ranging from various statistical techniques [8, 9, 10] to deep learning methods [11, 12, 13]. The sparse approximation based methods are outperforming existing techniques constrained to limited training data in terms of classification accuracy and easier implementation [14, 15, 16].…”

Section: Literature Surveymentioning

confidence: 99%

“…These higher level abstractions represent facial identities with outstanding stability. Therefore, the feature extracted from the pretrained deep learning models such as CNN shows remarkable performance [13]. VGGF, VGG16, VGG19 [2, 3], AlexNet, ResNet-50 [4] are few commonly used pre-trained CNN models for feature extraction.…”

Section: Introductionmentioning

confidence: 99%

Real Time Face Recognition with limited training data: Feature Transfer Learning integrating CNN and Sparse Approximation

Bajpai

Mishra

2021

Preprint

View full text Add to dashboard Cite

It is highly challenging to obtain high performance with limited and unconstrained data in real time face recognition applications. Sparse Approximation is a fast and computationally efficient method for the above application as it requires no training time as compared to deep learning methods. It eliminates the training time by assuming that the test image can be approximated by the sum of individual contributions of the training images from different classes and the class with maximum contribution is closest to the test image. The efficiency of the Sparse Approximation method can be further increased by providing high quality features as input for classification. Hence, we propose to integrate pre-trained CNN architecture to extract the highly discriminative features from the image dataset for Sparse classification. The proposed approach provides better performance even for one training image per class in complex environment as compared to the existing methods. Highlight of the present approach is the results obtained for LFW dataset with one and thirteen training images per class are 84.86% and 96.14% respectively, whereas the existing deep learning methods use a large amount of training data to achieve comparable results.

show abstract

“…Taigman et al [120] To investigate the long tail effect in deep facial recognition, Zhang et al [124] Several promising ideas have been explored to bring advances in CNNs, such as the use of proper activation [143] and various loss functions [15, 124-126, 130, 134, 136, 138, 141-142, 145, 148-152], the use of metric learning algorithms [127,38], normalization of features and weights [135,133,139], extraction of appearance variation features [123,129,147], use of multi-CNNs to extract features from various facial regions [116][117][118][119][120]122], and other ideas for the issue of imbalanced training data [131-132, 140, 144, 146]. The famous LFW benchmark results continue to climb as more deep face methods are introduced; for example, in pasting four years, the accuracy has been increased from 97.35% with DeepFace (2014) to 99.86% with COCO Loss (2017), as mentioned in Table 3.…”

Section: -Investigations Based On Alexnet Architecturementioning

confidence: 99%

Past, Present, and Future of Face Recognition: A Review

Adjabi¹,

Ouahabi²,

Benzaoui³

et al. 2020

Preprint

114

View full text Add to dashboard Cite

Face recognition is one of the most active research fields of computer vision and pattern recognition, with many practical and commercial applications including identification, access control, forensics, and human-computer interactions. However, identifying a face in a crowd raises serious questions about individual freedoms and poses ethical issues. Significant methods, algorithms, approaches, and databases have been proposed over recent years to study constrained and unconstrained face recognition. 2D approaches reached some degree of maturity and reported very high rates of recognition. This performance is achieved in controlled environments where the acquisition parameters are controlled, such as lighting, angle of view, and distance between the camera-subject. However, if the ambient conditions (e.g., lighting) or the facial appearance (e.g., pose or facial expression) change, this performance will degrade dramatically. 3D approaches were proposed as an alternative solution to the problems mentioned above. The advantage of 3D data lies in its invariance to pose and lighting conditions, which has enhanced recognition systems efficiency. 3D data, however, is somewhat sensitive to changes in facial expressions. This review presents the history of face recognition technology, the current state-of-the-art methodologies, and future directions. We specifically concentrate on the most recent databases, 2D and 3D face recognition methods. Besides, we pay particular attention to deep learning approach as it presents the actuality in this field. Open issues are examined and potential directions for research in facial recognition are proposed in order to provide the reader with a point of reference for topics that deserve consideration.

show abstract

Face recognition in unconstrained environment with CNN

Cited by 83 publications

References 32 publications

Past, Present, and Future of Face Recognition: A Review

Past, Present, and Future of Face Recognition: A Review

Real Time Face Recognition with limited training data: Feature Transfer Learning integrating CNN and Sparse Approximation

Past, Present, and Future of Face Recognition: A Review

Contact Info

Product

Resources

About