An Experiment to Find Gender & Age Classification Using CNNs

Dalali Arif; B Murali; Ipseeta Nanda

Dalali Arif Student, Department of Electronics and Communication, Koneru Lakshmaiah Education Foundation, Andhra Pradesh, India
B Murali Guide, Department of Electronics and Communication, Koneru Lakshmaiah Education Foundation, Andhra Pradesh, India
Ipseeta Nanda Faculty of Information Technology, Gopal Narayan Singh University, Jamuhar, Rohtas, Bihar, India

Abstract

Automatic age and gender characterization has gotten pertinent to an expanding number of utilizations, especially since the ascent of social stages and web-based media. In any case, execution of existing techniques on true pictures is still altogether missing, particularly when contrasted with the colossal jumps in execution as of late detailed for the connected assignment of face acknowledgment. In this paper we show that by learning portrayals using profound convolution neural organizations (CNN), a huge expansion in execution can be acquired on these undertakings. To this end, we propose a basic convolution net engineering that can be utilized in any event, when the measure of learning information is restricted. We assess our technique on the new Audience benchmark for age and sexual orientation assessment and show it to significantly beat present status of-the-workmanship strategies.

How to cite this article:
Arif D, Murali B. An Experiment to Find Gender and Age Classification Using CNNs. J Adv Res Appl Arti Intel Neural Netw 2021; 5(1): 22-26.

References

1. T. Ahonen, A. Hadid, and M. Pietikainen. Face description with local binary patterns: Application to face recognition Trans. Pattern Anal. Mach. Intell., 28(12):2037–2041, 2006.
2. W.-L. Chao, J.-Z. Liu, and J.-J. Ding. Facial age estimation based on label-sensitive learning and age-oriented regression. Pattern Recognition, 46(3):628–641, 2013.
3. E. Eidinger, R. Enbar, and T. Hassner. Age and gender estimation of unfiltered faces. Trans. on Inform. Forensics and Security, 9(12), 2014.
4. L. Wolf, T. Hassner, and Y. Taigman. Descriptor based methods in the wild. In post-ECCV Faces in Real-Life Images Workshop, 2008.
5. X. Zhuang, X. Zhou, M. Hasegawa-Johnson, and T. Huang. Face age estimation using patch-based hidden markov model supervectors. In Int. Conf. Pattern Recognition. IEEE, 2008.