• N.Indira
• Sana Rushitha Reddy
• Usirike Swetha Reddy
• Balamuri AmruthaSiree

Human pose approximation is a major problem in the stream of Computer Vision. Let us assume that we are capable of tracking a person’s daily activities and make an analysis in real time. The technology provides us many suggestions. Forecasting the parts of our body and positions of joints of an individual from the image or video is called as Post Estimation. Humans are flexible they are able to change their poses frequently. Now-a-days, this world is depending more on automation, so capturing all the activities are made in the surroundings using surveillances. It is difficult for an automatic computer to determine their poses for the analysis process. To analysis the human movement position Generative Adversarial Network (GAN) is used. GAN algorithm could help us to determine the exact pose of the human with highest efficiency rate, since this module uses generator and discriminator, in which the generator generates many reliable related images and the discriminator which tries to determine original and the fake. These two component make a efficient way for determining the exact pose the human and their gesture.

Human interaction, Communication, TM, Security.

"Human Activity Recognition using Generative Adversarial Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 4, page no.90-93, April-2021, Available :http://www.jetir.org/papers/JETIRES06020.pdf

"Human Activity Recognition using Generative Adversarial Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 4, page no. pp90-93, April-2021, Available at : http://www.jetir.org/papers/JETIRES06020.pdf