A Study of Deep Learning-Based Face Recognition Models for Sibling Identification

Goel, Rita; Mehmood, Irfan; Ugail, Hassan

doi:10.3390/s21155068

“…Since, YOLOv7 is a new state-of-the-art real-time object detection method (Wang, et al, 2023), hence we have used YOLOv7 to detect all the students present in the classroom and also further detect the face of each identi ed student for facial expression and eye state recognition or classi cation. And implement face recognition with the help of VGGFace (Simonyan & Zisserman, 2015; Parkhi, et al, 2015) model to identify each individual student's present in the classroom; the reason behind choosing VGGFace model for face recognition is that, many studies shows that VGGFace model has comparatively more accurate than other pre-trained models (Ghazi & Ekenel, 2016;Goel, et al, 2021;Grm, et al, 2018;Chandra & Reddy, 2020). And we performed Facial Expression classi cation using the ResNet neural network architecture (He, et al, 2016), because as per many studies the ResNet architecture could be a leading network architecture for image classi cation (Rahman, et al, 2020;Schieck, et al, 2023;Yang, et al, 2021;Mascarenhas & Agarwal, 2021).…”

Section: Introductionmentioning

confidence: 99%

Evaluation and Assessment of Teaching Quality and Students’ Performance using Machine Learning

Chakrabarti,

De

2024

Preprint

0

View full text Add to dashboard Cite

In quality education, we except good teaching. To improve the academic performance of a student it is important to monitor the students’ activities and attentiveness. The traditional way to do that is Examination and regular counselling; But an exam can’t conclude overall academic performance of a student, counselling is not enough to estimate the difficulties that students are facing in classroom, the level of attention of students in classroom and the favorite subjects of the students. And it’s also important to monitor the teachers’ teaching quality. In this work we proposed an AI model, by which we can analysis the students’ emotion, activities and attentiveness in the classroom with the help of high-resolution configured camera in the classroom. The model detects the facial expressions of each student in classroom to recognize the emotion of each student; and detect the eyes, lips and head movement of each student to recognize the activities of the students. And analyze these recognized activities for each student to classify the attentiveness status (attentive or inattentive) of each student. And an overall activities and emotion of all the students will be analyzed to evaluate the teacher’s teaching quality at the class – If most of the students are inattentive or not happy, means the teacher’s teaching methods is not helping the students and vice versa. The proposed machine leaning model not only analyze performance of the student and teaching method, it also alerts the teachers about the unwanted activities of the student in the classroom in real-time.

show abstract

“…1], hence we have used YOLOv7 to detect all the students present in the classroom and also further detect the face of each identified student for facial expression and eye state recognition or classification. And implement face recognition with the help of VGGFace [2,3] model to identify each individual student's present in the classroom; the reason behind choosing VGGFace model for face recognition is that, many studies shows that VGGFace model is comparatively more accurate than other pre-trained models for face recognition [4,5,6,7]. And we performed Facial Expression classification using the EfficientNet neural network architecture [8,9].…”

Section: Introductionmentioning

confidence: 99%

Evaluation and Assessment of Teaching Quality and Students’ Performance using Machine Learning

Chakrabarti,

De

2024

Preprint

0

View full text Add to dashboard Cite

In quality education, we except good teaching. To improve the academic performance of a student it is important to monitor the students’ activities and attentiveness. The traditional way to do that is Examination and regular counselling; But an exam can’t conclude overall academic performance of a student, counselling is not enough to estimate the difficulties that students are facing in classroom, the level of attention of students in classroom and the favorite subjects of the students. And it’s also important to monitor the teachers’ teaching quality. In this work we proposed an AI model, by which we can analysis the students’ emotion, activities and attentiveness in the classroom with the help of high-resolution configured camera in the classroom. The model detects the facial expressions of each student in classroom to recognize the emotion of each student; and detect the eyes, lips and head movement of each student to recognize the activities of the students. And analyze these recognized activities for each student to classify the attentiveness status (attentive or inattentive) of each student. And an overall activities and emotion of all the students will be analyzed to evaluate the teacher’s teaching quality at the class – If most of the students are inattentive or not happy, means the teacher’s teaching methods is not helping the students and vice versa. The proposed machine leaning model not only analyze performance of the student and teaching method, it also alerts the teachers about the unwanted activities of the student in the classroom in real-time.

show abstract

“…Nowadays, for the purpose of such human detection, a machine learning-based face recognition model [6,7] can be a good solution. However, there are some considerations that make it difficult to directly apply typical face recognition models to the proposed system.…”

Section: Introductionmentioning

confidence: 99%

A Regional Brightness Control Method for a Beam Projector to Avoid Human Glare

Jeon,

Lee

2024

Applied Sciences

0

View full text Add to dashboard Cite

In this study, we proposed a system to reduce the speaker’s suffering from the strong light of a beam projector by applying regional brightness control over the screen. Since the original image and the projected one on the screen are quite different in area, brightness, and color, the proposed system first transforms them so that they have the same area and similar color tone. Then, to accurately determine the difference between those images, we have introduced a SSIM map, which is a perception-based method of measuring image similarity. Accordingly, an image segmentation model is used to determine the speaker’s silhouette from the SSIM map. We applied a couple of well-trained segmentation models, such as Selfie and DeepLab-v3, provided with MediaPipe. The experimental results showed the operability of the proposed system and that it determines most of a lecturer’s body area on the screen. To closely evaluate the system’s effectiveness, we have measured error rates consisting of false-positive and false-negative errors in the confusion matrix. With the measured results, the error rates appeared so insignificant and stable that the proposed system provides a practical effect for the speakers, especially in the case of applying DeepLab-v3. With the results, it is implied that an accurate segmentation model can considerably elevate the effectiveness of the system.

show abstract

A Study of Deep Learning-Based Face Recognition Models for Sibling Identification

Cited by 27 publications

References 35 publications

Evaluation and Assessment of Teaching Quality and Students’ Performance using Machine Learning

Evaluation and Assessment of Teaching Quality and Students’ Performance using Machine Learning

Evaluation and Assessment of Teaching Quality and Students’ Performance using Machine Learning

A Regional Brightness Control Method for a Beam Projector to Avoid Human Glare

Contact Info

Product

Resources

About