Please use this identifier to cite or link to this item: http://10.1.7.192:80/jspui/handle/123456789/11891
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKaur, Bhupinder-
dc.date.accessioned2023-08-18T07:25:29Z-
dc.date.available2023-08-18T07:25:29Z-
dc.date.issued2023-06-01-
dc.identifier.urihttp://10.1.7.192:80/jspui/handle/123456789/11891-
dc.description.abstractThe ancient times, during 5000 years ago, the Indus-Sarasvati culture in ancient India created the yoga practise. The term "yoga" refers to a close connection and integration of the mind and the body. Through asana, meditation, and other practises, it is utilised to keep the body and mind in balance throughout all of life's ups and downs. Due of the growing stress levels in the modern lifestyle, yoga has recently attracted interest on a global scale. There are several ways to practise yoga. Many people choose self-learning in fast-paced environments because the aforementioned resources might not always be accessible. In our method, as input, we use the personalised yoga dataset. The system uses various deep learning techniques to determine whether the input yoga stance image is correct or incorrect. Results display metrics for performance including accuracy, error rate, and comparison graph. There are several ways to accomplish this; however, the method I employ begins by using a CNN to process the incoming image and ANN classifier that has been trained to seek for humans. The pose estimation network looks for limbs and joints with training when it recognises human body positions. The user can then view the image on the computer using markers that designate various bodily components. However, we did the implementation with the above mentioned CNN and ANN Machine learning algorithm the 2d detection of yoga poses by giving feedback to improve position of the body parts and detect the pose with classification. We need a design and implementation on the basis of real-time detection and classification of the Yoga pose performed by the user. Hence, we will inculcate the previous algorithm and work on the real-time 3d pose estimation and correction by implementing yoga pose detection System which is designed and developed to recognize yoga stances and respond with a customized response to help users improve their postures. Our system will detect various yoga poses, namely Chair pose (Utkatasana), Cobra pose (Bhujangasana), Dog Pose (Adho Mukha Svanasana), Shoulder Stand Pose (Sarvangasana), Triangle Pose (Trikonasana), Tree Pose (Vrikshasana), Warrior Pose (Virbhadrasana).en_US
dc.publisherInstitute of Technologyen_US
dc.subjectComputer 2021en_US
dc.subjectProject Report 2021en_US
dc.subjectComputer Project Reporten_US
dc.subjectProject Reporten_US
dc.subject21MCEen_US
dc.subject21MCEDen_US
dc.subject21MCED09en_US
dc.subjectCE (DS)en_US
dc.subjectDS 2021en_US
dc.title3D Yoga Pose Estimation and Correctionen_US
dc.typeDissertationen_US
Appears in Collections:Dissertation, CE (DS)

Files in This Item:
File Description SizeFormat 
21MCED09.pdf21MCED09713.31 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.