Yoga Pose Classification Using Input Image Dataset and Fine tuning it on MobileNet Model with Adadelta Optimizer

Yoga has been the focus of several scientific studies in a variety of fields because of its potential advantages for both physical and mental health. Yoga's impact on a range of health issues, such as cardiovascular health, chronic pain, mental health disorders, and musculoskeletal ailments, have been the subject of research. Studies have looked at how yoga affects controlling chronic illnesses, improving flexibility and strength, lowering stress, improving sleep quality, and boosting general wellbeing. Yoga's impacts are being studied scientifically in order to better understand its potential advantages, develop evidence-based practises, and integrate yoga into other fields such as wellness initiatives and healthcare. The objective of this study is to categorise yoga positions using the input photographs. Data augmentation using Keras is also put into practise. Using the sustainable public accessible “MobileNet” model that has been trained on the “imagenet” dataset, a model implementation via “Transfer learning” is demonstrated, and the results indicate 96% validation accuracy when the improvement is carried out with fine tuning and how it can create an environmental impact.