• Suteja Patil
• Prof.Abhilasha Kulkarni
• Radhika Khiste
• Shreeya Tahasildar
• Shreeya Tambolkar

In this paper, we propose an automatic age and gender recognition system from a live-video stream. Haar cascade classifier is used for face detection while for age and gender recognition,VGG16 is used. VGG-16 is a famous deep-Convolutional Neural Networks(CNN) architecture.VGG-16 network is trained on ImageNet dataset which has over 14 million images and 1000 classes, and achieves 92.7% top-5 accuracy. It surpasses AlexNet network by replacing large filters of size 11 and 5 in the first and second convolution layers with small size 3x3 filters. Transfer learning is implemented for iterative transfer of knowledge and better recognition with greater accuracy. IMDB-WIKI dataset is used for training and testing of the model which is the largest dataset of human faces with gender, name and age information. This real-time prediction model is characterized by more accuracy when compared to the publicly available methods. As face analysis is a challenging task because of variations in images having different postures, lighting ,angles and expressions, this model successfully overcomes all the barricades and proves to be the most accurate while using less computational resources.

"Video-based Human Age and Gender Recognition using Deep Convolutional Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 4, page no.1090-1093, April-2020, Available :http://www.jetir.org/papers/JETIR2004344.pdf

"Video-based Human Age and Gender Recognition using Deep Convolutional Neural Networks", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 4, page no. pp1090-1093, April-2020, Available at : http://www.jetir.org/papers/JETIR2004344.pdf