Detecting Disguised Faces with Transfer Learning
Dr. A. Prabhu1, Daphedari Kishan Prasad2, Adabala Taraka Rama Venkata Sai Hanuman3 and Abhinav Joel T4
1 Associate Professor, Department of Computer Science and Engineering, CMR Technical Campus, Medchal, Telangana, India.
2 UG Student, Department of Computer Science and Engineering, CMR Technical Campus, Medchal, Telangana, India.
3 UG Student, Department of Computer Science and Engineering, CMR Technical Campus, Medchal, Telangana, India.
4 UG Student, Department of Computer Science and Engineering, CMR Technical Campus, Medchal, Telangana, India.
ABSTRACT:
In general, a human being has the memory power due to which he/she will be able to remember whatever they have seen but as the technology is increasing the advancement is increasing in such a way that now computer is also able to recognize theX faces from its memory but in order to differentiate them, we need more advancement which leads to the development of Machine learning. Machine learning concepts developed by Arthur Manuel. There have been many techniques used over the past decade to determine the identity of a person's face, such as Eigenfaces and Principal Component Analysis (PCA), to Convolutional Neural Networks (CNN) to ensure the ability to recognize faces has become further and further. An approach to machine learning called transfer learning involves creating a model of the first training task, then testing it using the model. The difference between transfer learning and traditional machine learning is that translation involves using a pre-trained model in order to start a secondary task using the initial model. It is expected that this paper will contribute to the field of image classification by using Machine Learning algorithms to solve the problem. Transfer learning significantly improves the performance of VGG models Based on these results, we conclude that VGG Models are the best choice for recognizing faces using ImageNet weights
Key words: Image, Memory, Machine Learning, Techniques, CNN models, Face Recognition, Deep Learning, Transfer Learnings
