• Deepak Bishnoi
• Dr. Kapil Kaswan

Today “Fit India” movement is on its heights and also a healthy body is the need of the hour and everybody wants to achieve it. For which they find yoga as a perfect and cheaper option. But as a fresher everyone needs a trainer to guide them with yoga poses. But in yoga no one can serve as a perfect trainer due to lack of excellence and knowledge thus automatic detection may help in doing proper yoga exercises. Work out must be done in correct way or correct posture. Any change in the posture may adversely affect the body so we need a trainer for proper work out. In automatic detection of yoga postures, the main intention is practicing yoga without a trainer. So, my motivation is those who either can’t afford trainer or shy of doing yoga in the presence of others. This research is an intent to help those who can’t find a trainer to tell them about accuracy in pose. With this research they have to click their images in different poses and compare them with reference images by image detectors. Following are the main outlines of current research: Total 18 poses were detected under this research work these 18 are the most common yoga poses which are usually performed for fitness and better physique. Both GFD and ANN are used for achievement of research objectives that is detection and comparison of yoga poses. The accuracy by GFD method is more than 93%. Along with shape descriptor, the key-point angle method serve for maximum precision of the results. 600 database images are created for covering all the 18 yoga postures. HoG and SOBEL is used as a shape descriptor.

Yoga poses, detection, shape, key features, descriptor.

"AUTO-DETECTION OF DIFFERENT YOGA POSES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.311-315, June 2019, Available :http://www.jetir.org/papers/JETIR1907D47.pdf

"AUTO-DETECTION OF DIFFERENT YOGA POSES", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp311-315, June 2019, Available at : http://www.jetir.org/papers/JETIR1907D47.pdf