Modern Face Recognition with Deep Learning

Thilaga, P. Jothi; Khan, Baber; Jones, Andrew; Kumar, N. Krishna

doi:10.1109/icicct.2018.8473066

Cited by 12 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This concept is very user-friendly and reduces a lot of manpower which goes along with the current trend of automation, there is no chance for human error and it is very easy to add more features into the current system. Image recognition or machine learning [6,15] in general needs a lot of data to give viable results. This would be possible by implementing this project with the cloud which consists of an array of data that can be extracted and used to get better results.…”

Section: Discussionmentioning

confidence: 99%

A Proposed System for Security Surveillance to Detect Human Intrusion

2019

IJITEE

View full text Add to dashboard Cite

Nowadays, Security is one of the, if not the biggest priority in places such as public speaking venues, restricted areas and all important places of any country. Moreover, military conflicts & terrorism has increased recently. Monitoring these areas is substantial, but currently, monitoring is dependent upon technology and manpower. Since there is always the potential of having a human error that may be caused due to several reasons, to counter this issue, this research paper proposed an application to reduce human error. The primary feature of this application is a surveillance system that detects the face of a human, identifies the personality and display their information to possibly prevent human intrusion..

show abstract

Section: Discussionmentioning

confidence: 99%

A Proposed System for Security Surveillance to Detect Human Intrusion

2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…Jothi Thilaga.P, ArshathKhan.B, Jones.A. A, Krishna Kumar N [3] This research visualizes the Facial recognition systems have limited accuracy due to occlusions, pose, and illumination changes, but can be improved with the use of hog descriptors and deep learning techniques. A Python-based application is being developed to faces in various conditions to recognize.…”

Section: Literature Reviewmentioning

confidence: 96%

Dusky Detect: Facial Detection in Low Light Using CNN

Kumar¹,

Parvathi²,

Manasa³

et al. 2023

Int. J. Res. Publ. Rev.

View full text Add to dashboard Cite

At present, Identification of faces through image processing is a crucial task in the fields of Machine Learning and Artificial intelligence. Luminescence plays a key role in the face detection. Nowadays, detecting the faces of different people present in different lightings and different times of day has become strenuous.Dusky face is sticky to predict using traditional methods. In order to find a solution, this project proposes various classifiers and Convolutional Neural Networks to detect the face in Abnormal lighting situations. This idea provides the process and technique of the face detection in low light using CNN. The proposed idea in this project will improve the accuracy and facial recognition in different light intensities using the combinations of various facial recognition techniques and classifiers and also the CNN for obtaining a better model with higher performance metrics.

show abstract

“…A skeleton detection module is used to detect all subjects present in each frame [16] and label their bodies to determine their heads Region of Interest (RoI). A facial recognition module is then applied to identify the main subject among all subjects [17]. The skeleton detection module determines the RoI of the pair of hands of the detected main subject.…”

Section: Background and Related Workmentioning

confidence: 99%

“…For the facial recognition model, we used a condition-free facial recognition library (no illumination changes or occlusions) with 99.38% accuracy that needs just one sample image for training [17]. The facial recognition model creates the landmark based on the provided single face image from the main subject and uses it to detect the main subject's face among the detected head RoIs from the skeleton detection module.…”

Section: Facial Recognitionmentioning

confidence: 99%