Virtual Fitness Trainer using Artificial Intelligence

Abstract

An AI fitness trainer is a computer application that utilizes the capabilities of Python, OpenCV, and MediaPipe to guide users through physical fitness routines. The application uses computer vision techniques provided by OpenCV to track the user\'s movements and provide feedback on form and technique. MediaPipe is used to process the video data and provide real-time analysis. The application also utilizes machine learning algorithms to provide personalized fitness recommendations and progress tracking. The combination of these technologies provides a highly interactive and effective way for users to improve their physical fitness.

Introduction

I. INTRODUCTION

Artificial Intelligence (AI) has the potential to revolutionize the way we approach physical fitness, offering personalized and interactive experiences that go beyond what is possible with traditional fitness training methods. One example of this is an AI fitness trainer that combines the power of Python, OpenCV, and MediaPipe to provide users with real-time feedback and guidance.

The use of computer vision techniques provided by OpenCV allows the application to track the user's movements, providing feedback on form and technique in real-time. MediaPipe processes the video data, providing fast and efficient analysis to support the application's real-time feedback capabilities.

In addition to real-time feedback, the AI fitness trainer utilizes machine learning algorithms to provide personalized fitness recommendations based on the user's abilities and goals. This allows the application to evolve and adapt to the user's needs over time, providing a highly personalized fitness experience. With the capability to track the user's progress, the application enables users to monitor their achievements and determine the extent of their progress towards achieving their fitness objectives.

The combination of these technologies provides a unique and highly interactive way for users to improve their physical fitness. The AI fitness trainer offers real-time feedback and guidance that is both accurate and personalized, making it an effective tool for people of all fitness levels. Whether you're a beginner just starting out or an experienced fitness enthusiast looking to take your training to the next level, this application provides the tools you need to achieve your goals.

In conclusion, the AI fitness trainer built using Python, OpenCV, and MediaPipe represents a major step forward in the field of fitness training, offering a highly personalized and interactive experience that is not possible with traditional methods. Whether you're looking to get fit, lose weight, or simply improve your overall health and wellness, this application provides the guidance and support you need to achieve your goals.

A. Objective

The objective of the AI fitness trainer built using Python, OpenCV, and MediaPipe is to provide a highly interactive and personalized fitness experience for users. This application aims to address the limitations of traditional fitness training methods by utilizing advanced technologies to offer real-time feedback and guidance.

The use of computer vision techniques provided by OpenCV allows the application to track the user's movements and provide immediate feedback on form and technique. MediaPipe processes the video data, providing fast and efficient analysis to support the real-time feedback capabilities of the application.

Machine learning algorithms are used to provide personalized fitness recommendations based on the user's abilities and goals.

In summary, the AI fitness trainer built using Python, OpenCV, and MediaPipe is designed to offer users a unique and highly effective way to improve their physical fitness.

II. LITERATURE REVIEW

"AI-powered Fitness Training: A Review of the Literature" is a paper that presents an outline of the current studies on the application of artificial intelligence in fitness training. The authors conduct a comprehensive search of the literature and analyze the existing studies to provide an overview of the field.

The authors found that AI-powered fitness training has the potential to provide personalized training programs, improve the accuracy of physical activity recognition, and provide real-time feedback and motivation to users. The authors also highlight the challenges of using AI in fitness training, such as privacy and data security concerns, and the need for additional research to evaluate the effectiveness of AI-powered fitness training.

In conclusion, the authors suggest that AI-powered fitness training has the potential to revolutionize the field of fitness and wellness by providing personalized and effective training programs. Nevertheless, they underscore the necessity for further research to comprehensively grasp the potential advantages and drawbacks of this technology.

Overall, this literature review provides a valuable overview of the field of AI-powered fitness training and highlights the need for further research in this area. [1]

The paper "Artificial Intelligence-based Personal Fitness Trainer" by Dr. S. M. Patil et al. provides a review of the existing research on the use of artificial intelligence in personal fitness training. The authors conducted a comprehensive search of the literature and analyzed existing studies to provide an overview of the field.

The authors found that AI-based personal fitness trainers have the potential to provide personalized training programs, real-time feedback and motivation to users, and improve the accuracy of physical activity recognition. The authors also discuss the challenges of using AI in fitness training, such as data privacy and security concerns, and the need for additional research to evaluate the effectiveness of AI-based personal fitness training.

The authors also highlight the current trends in AI-based personal fitness training, including the use of wearable devices and mobile applications, and the integration of machine learning and deep learning algorithms. They also provide an overview of the existing systems and applications for AI-based personal fitness training.

In conclusion, the authors suggest that AI-based personal fitness training has the potential to revolutionize the field of fitness and wellness by providing personalized and effective training programs. However, it is also emphasized that more research is required to gain a complete understanding of the technology's potential advantages and limitations.

Overall, this literature review provides a valuable overview of the field of AI-based personal fitness training and points to the need for further research in this area. The authors provide a comprehensive analysis of the existing studies and highlight the current trends in the field, making it a valuable resource for those interested in AI-based personal fitness training. [2]

The paper "AI Fitness Coach at Home using Image Recognition" by Ji et al. focuses on the use of image recognition in AI-powered fitness coaching. The authors describe a system that uses image recognition to provide users with personalized exercise guidance and feedback in the comfort of their own homes.

The authors found that image recognition can improve the accuracy of physical activity recognition and provide real-time feedback and motivation to users. They also discuss the challenges of using image recognition in fitness coaching, such as the need for large training datasets and the limitations of the technology in recognizing more complex movements.

The authors present a prototype system that uses image recognition to provide personalized exercise guidance and feedback to users in real-time. The system includes a camera that captures images of the user's movements, which are then analyzed using machine learning algorithms to provide accurate and personalized feedback. The authors also provide an evaluation of the system, which shows that it can accurately recognize a range of exercises with a high degree of accuracy.

In conclusion, the authors suggest that AI-powered fitness coaching using image recognition has the potential to revolutionize the field of fitness and wellness by providing personalized and effective training programs.

Overall, this paper provides a valuable contribution to the field of AI-powered fitness training by exploring the use of image recognition in this context. The authors present a prototype system that demonstrates the feasibility of using image recognition for personalized fitness coaching and provide an evaluation of its performance, making it a valuable resource for those interested in this area. [3]

The paper "AI Fitness Trainer" by Kashish Jain et al. provides a review of the existing research on the use of artificial intelligence in personal fitness training. The authors conduct a comprehensive search of the literature and analyze existing studies to provide an overview of the field. The authors found that AI-powered personal fitness trainers have the potential to provide personalized training programs, real-time feedback and motivation to users, and improve the accuracy of physical activity recognition. They also discuss the challenges of using AI in fitness training, such as data privacy and security concerns, and the need for additional research to evaluate the effectiveness of AI-based personal fitness training.

The authors also highlight the current trends in AI-based personal fitness training, including the use of wearable devices and mobile applications, and the integration of machine learning and deep learning algorithms. They provide an overview of the existing systems and applications for AI-based personal fitness training and their limitations.

In conclusion, the authors suggest that AI-based personal fitness training has the potential to revolutionize the field of fitness and wellness by providing personalized and effective training programs. However, they also emphasize the need for additional research to fully understand the potential benefits and limitations of this technology.

Overall, this literature review provides a valuable overview of the field of AI-based personal fitness training and highlights the need for further research in this area. The authors provide a comprehensive analysis of the existing studies and highlight the current trends in the field, making it a valuable resource for those interested in AI-based personal fitness training. [4]

The paper "How to Increase Sport Facility Users’ Intention to Use AI Fitness Services: Based on the Technology Adoption Model" by Chin et al. focuses on the factors that influence sport facility users' intention to use AI fitness services. By employing the Technology Adoption Model (TAM), the authors analyze how the perceived usefulness, perceived ease of use, and perceived enjoyment contribute to the adoption of AI-based fitness services. The authors discovered that the users' intention to utilize AI fitness services is significantly influenced by their perceived usefulness, perceived ease of use, and perceived enjoyment of such services in sports facilities. They also found that the factors affecting perceived usefulness include the accuracy of feedback, the provision of personalized training programs, and the availability of real-time monitoring and motivation.

To gather data on the users' outlook towards AI fitness services, the authors administered a survey to the sport facility users. The survey findings indicated a positive correlation between the users' intention to use AI fitness services and their perceived usefulness, perceived ease of use, and perceived enjoyment of these services.

In conclusion, the authors suggest that sport facility managers can increase the adoption of AI fitness services by focusing on perceived usefulness, perceived ease of use, and perceived enjoyment. They also suggest that sport facility managers can increase perceived usefulness by providing accurate feedback, personalized training programs, and real-time monitoring and motivation.

By analyzing the influence of perceived usefulness, perceived ease of use, and perceived enjoyment on the adoption of AI fitness services by sport facility users, this paper offers important perspectives. The paper's utilization of the Technology Adoption Model enhances the understanding of AI fitness services and makes it a valuable resource for individuals keen on exploring these services and their acceptance by sport facility users. [5]

The paper "Object Detection using OpenCV and Python" by Sharma et al. focuses on the development of a real-time object detection system using the OpenCV library and Python programming language. The authors describe the use of various computer vision algorithms and techniques, such as Haar cascades, HOG (Histogram of Oriented Gradients), and deep learning, for object detection.

The authors evaluate the performance of their object detection system by comparing its accuracy and speed with other existing systems. The results of the evaluation showed that the system developed by the authors achieved a high level of accuracy and speed, making it a suitable choice for real-time object detection applications.

The authors also discuss the limitations and challenges of object detection systems and provide suggestions for future work. For example, they highlight the need for improving the accuracy of object detection in complex scenes, such as crowded scenes and scenes with occlusions.

In conclusion, the authors present a real-time object detection system using the OpenCV library and Python programming language. The system provides a high level of accuracy and speed, making it a suitable choice for real-time object detection applications. The authors also discuss the limitations and challenges of object detection systems and provide suggestions for future work, making this paper a valuable resource for those interested in computer vision and object detection. [6]

The paper "Pose Estimation and Correcting Exercise Posture" by Kanase et al. focuses on the development of a system for estimating and correcting exercise posture. The authors describe how the system uses computer vision techniques, such as pose estimation and deep learning, to estimate the posture of an individual during exercise and provide feedback on how to correct their posture.

The authors evaluate the performance of their posture estimation and correction system by conducting experiments on a dataset of individuals performing exercises. The results of the evaluation showed that the system was able to accurately estimate the posture of individuals and provide meaningful feedback on how to correct their posture.

The authors also discuss the potential benefits of using the system in a fitness training setting. For example, they highlight how the system can help individuals improve their posture during exercise, reduce the risk of injury, and improve the effectiveness of their workout.

In conclusion, the authors present a system for estimating and correcting exercise posture using computer vision techniques. The system was shown to be accurate in estimating posture and providing meaningful feedback, making it a valuable tool for individuals looking to improve their posture during exercise. The authors also discuss the potential benefits of using the system in a fitness training setting, making this paper a valuable resource for those interested in computer vision and fitness training. [7]

The paper "AI-based Workout Assistant and Fitness guide" by Taware et al. focuses on the development of a system for providing workout assistance and fitness guidance to individuals. The authors describe how the system uses artificial intelligence techniques, such as computer vision and machine learning, to provide customized workout plans and real-time feedback on exercise performance.

The authors evaluate the performance of their workout assistant system by conducting experiments on a dataset of individuals performing exercises. The results of the evaluation showed that the system was able to accurately track exercise performance and provide meaningful feedback on how to improve.

The authors also discuss the potential benefits of using the system in a fitness training setting. For example, they highlight how the system can help individuals follow a personalized workout plan, monitor their progress, and receive real-time feedback on their performance.

In conclusion, the authors present a system for providing workout assistance and fitness guidance using artificial intelligence techniques. The system was shown to be effective in tracking exercise performance and providing meaningful feedback, making it a valuable tool for individuals looking to improve their fitness. The authors also discuss the potential benefits of using the system in a fitness training setting, making this paper a valuable resource for those interested in artificial intelligence and fitness training. [8]

The paper "The effectiveness of a personalized virtual fitness trainer in teaching physical education by applying the artificial intelligent algorithm" by Nur Azlina Mohamed Mokmin investigates the effectiveness of a customized virtual fitness trainer for imparting physical education. The authors' objective is to assess the virtual fitness trainer's efficiency in enhancing user engagement and knowledge acquisition.

The study was conducted with a group of physical education students who used the virtual fitness trainer for a period of six weeks. The authors found that the virtual fitness trainer was effective in improving user engagement, as measured by increased time spent using the system and high user satisfaction scores. Additionally, the virtual fitness trainer was effective in enhancing the students' knowledge of physical education, as measured by pre- and post-intervention assessments.

The authors also discuss the potential benefits of using a personalized virtual fitness trainer in teaching physical education. For example, they highlight the potential for increased engagement and motivation, as well as the ability to provide individualized feedback and guidance.

In conclusion, the authors present evidence that a personalized virtual fitness trainer can be an effective tool for teaching physical education. The results of the study suggest that the virtual fitness trainer can improve user engagement and knowledge acquisition, making it a valuable resource for educators and students alike. The authors also discuss the potential benefits of using a personalized virtual fitness trainer, making this paper a valuable resource for those interested in the intersection of artificial intelligence, physical education, and fitness training. [9]

The paper "PifPaf: Composite Fields for Human Pose Estimation" by Sven Kreiss, Lorenzo Bertoni, and Alexandre Alahi focuses on the development of a new framework for human pose estimation using artificial intelligence (AI) techniques. The authors present PifPaf, a new method that uses composite fields to accurately estimate the human body's keypoints. The method is evaluated using several benchmark datasets and compared with existing state-of-the-art methods.

PifPaf shows promising results, with improved accuracy over previous methods. The use of composite fields allows the algorithm to effectively handle occlusions, which are a common problem in human pose estimation. Additionally, PifPaf is fast and scalable, making it suitable for real-time applications.

The authors conclude that PifPaf is a promising method for human pose estimation, with the potential to be applied in various fields, including fitness training. The accuracy and speed of the algorithm make it a valuable tool for developing AI-powered fitness trainers.

Overall, the "PifPaf: Composite Fields for Human Pose Estimation" paper provides important contributions to the field of human pose estimation and has significant implications for the development of AI-powered fitness trainers. The authors' innovative approach to solving the challenges of human pose estimation offers a new perspective and direction for future research in the field. [10]

The paper "OpenCV for Computer Vision Applications" by Naveenkumar and Ayyasamy Vadivel presents an overview of the use of OpenCV, an open-source computer vision library, for various computer vision applications. The authors highlight the key features of OpenCV and explain how it can be used for various computer vision tasks such as image processing, object detection, and pose estimation. They provide a brief history of the development of computer vision and how OpenCV fits into this development. The authors also present a few use cases of OpenCV, including its use in controlling robotic systems and surveillance systems.

The paper focuses on the ease of use and accessibility of OpenCV, which has made it a popular choice among computer vision developers. The authors note that OpenCV has a large community of developers and users, which contributes to its constant improvement and expansion. The authors also point out the wide range of algorithms and techniques available in OpenCV, including deep learning-based techniques, which makes it possible to tackle complex computer vision problems.

Overall, the paper gives a concise overview of OpenCV and its applications in computer vision. It provides a good starting point for researchers and developers who are interested in using OpenCV for their computer vision projects. [11]

The paper "A Study on the Feasibility of AI-based Personalized Physical Training" by H. Lee, J. Kim, and J.Y. Kim investigates the potential of using artificial intelligence (AI) to provide personalized physical training. The authors aim to explore the feasibility of an AI-based physical training system, including the development of a framework for such a system and the assessment of its performance and user satisfaction.

The authors used a combination of AI techniques, such as machine learning and computer vision, to develop a framework for the AI-based physical training system. This system was then evaluated in terms of its performance and user satisfaction. The results showed that the AI-based physical training system was able to provide personalized training plans that were effective and well-received by users.

The paper highlights the potential benefits of using AI to provide personalized physical training, including the ability to provide customized training plans based on an individual's physical abilities and goals. The authors suggest that such a system could help to improve overall fitness levels, as well as increase motivation and engagement in physical activity.

Overall, the results of the study support the feasibility of using AI to provide personalized physical training and suggest that further research in this area could lead to significant advances in the field of fitness and health. [12]

III. METHODOLOGY

The AI fitness trainer uses a machine learning algorithm such as a convolutional neural network (CNN) to analyze video frames captured by a camera and detect the key points of a person's body (such as joints and limbs) using the pre-trained models in Mediapipe. The OpenCV library can then be used to track the movement of these body parts over time and provide feedback on the person's form and technique.

A. Data Collection

Data collection is a critical part of the AI training process. The data you collect will be used to train machine learning models to make predictions or classifications based on the patterns and relationships found in the data. Mediapipe and OpenCV are both popular libraries used in computer vision for data collection and processing. You will need to connect a camera to your computer and configure it to capture the images you need. You can use OpenCV to capture images or video frames from the camera. These images can then be processed using Mediapipe to extract features like hand gestures and body pose. Once you have collected the data, you may need to pre-process it to clean or transform it. This could include cropping the images, resizing them, or converting them to grayscale. To use the data for training a machine learning model, you will need to label it. This involves assigning each image or frame to a specific category or class. Finally, you need to store the labelled data in a format that can be used for training a machine learning model, such as CSV or TFRecord.

B. Pre-processing

Data processing is an important step in preparing data for machine learning models, and Mediapipe and OpenCV can be useful tools for this process. The first step is to collect the data that you want to use for training your AI model. This could include images, videos, or other types of visual data. Mediapipe and OpenCV can be used to clean the data by removing invalid or redundant samples, or filling in missing data.

Once the data is clean, you can use Mediapipe and OpenCV to pre-process it by normalizing the data, augmenting it, and extracting features. Normalization can involve scaling the data to a standard range or distribution, while augmentation involves generating new samples from existing data, such as rotating or flipping images. Feature extraction involves using Mediapipe to extract specific features from the data, such as hand gestures or body pose.

C. Training

Training an AI trainer using Mediapipe and OpenCV involves a series of iterative steps, where you adjust the model's parameters and evaluate its performance until it achieves the desired level of accuracy and reliability. Once the data is pre-processed, split it into training, validation, and testing sets. You could use a convolutional neural network (CNN) for image classification tasks. Use the training set to train the machine learning model using the chosen algorithm. This involves feeding the pre-processed data into the model and adjusting its internal parameters to improve its performance. After training the model, use the validation set to fine-tune the model's parameters and improve its accuracy.

This involves making small adjustments to the model and testing its performance on the validation set to see if the changes have improved its performance. Once the model has been fine-tuned, evaluate its performance on the testing set to see how well it can make predictions or classifications on new data. This involves comparing the model's predictions to the actual values in the testing set and calculating performance metrics, such as accuracy, precision, recall, and F1-score. Finally, if the model performs well on the testing set, deploy it to make predictions or classifications on new data.

D. Validation

Validation is used to assess the accuracy and reliability of the model's predictions or classifications on new data. This is important because the model may have learned patterns or features in the training data that are not representative of the real-world data it will encounter in practice.

By using validation data, which is separate from the data used for training the model, we can test whether the model is able to generalize well and make accurate predictions or classifications on new, unseen data.

Validation aids in avoiding overfitting, which arises when a model is excessively intricate and closely matches the training data, causing suboptimal performance on novel data. By evaluating the model's performance on validation data, we can detect whether the model is overfitting and adjust its parameters or algorithms to improve its generalization performance.

IV. SCREENSHOTS

Below are a few sample screenshots which explains the working.

In Fig.3; the process of selecting a workout option is simple and easy. The user is presented with multiple choices and they can simply choose the one they prefer by clicking on it. Once the option is selected, the user is directed to the next page where they can begin their workout session. This process eliminates the need for complicated navigation or extensive searching for the desired workout option. The user can quickly and conveniently access the workout they desire with just a few clicks.

In Fig.4; to make it easier for users to keep track of their workout progress, a feature has been included that allows them to count the number of times a specific workout has been done. So just enter the number of times the workout needs to be performed and click the submit button. This eliminates the need for users to manually keep track of their workout count, as the system will do it for them. With this feature, users can simply focus on their workouts without the added stress of trying to remember how many times they have performed a particular exercise. This makes the workout experience more enjoyable and convenient for users, as they can now easily monitor their progress with just a few clicks.

In Fig.5; upon submitting the workout count, the camera is activated and keypoints are located on the body based on the exercise that has been selected. As the user performs the workout, the system automatically counts the number of repetitions and the calories burned, and displays the information on the screen in real-time. This allows users to easily track their progress and stay motivated during their workout session. With the system's ability to simultaneously count the number of repetitions and calories burned, users can get a comprehensive overview of their workout performance and adjust their routine as needed to achieve their fitness goals.

In Fig.6; after the user completes their workout, the camera is automatically turned off and they are directed to a page that displays their workout results. This page shows the count of reps that were completed and the total number of calories burned during the workout session. It's important to note that the results are only displayed once the workout has been completed, ensuring that the user receives accurate and reliable information about their performance. With this feature, users can easily keep track of their workout progress and monitor their fitness goals over time.

V. RESULTS

An AI fitness trainer that uses Mediapipe and OpenCV can provide several benefits to users. This type of fitness trainer uses advanced machine learning algorithms and computer vision techniques to track a person's movements during exercise routines and provide real-time feedback on their form and technique. Some of the specific results that can be achieved through the use of an AI fitness trainer with Mediapipe and OpenCV include:

1. Improved form and Technique: One of the main benefits of using an AI fitness trainer that uses Mediapipe and OpenCV is that it can help users improve their form and technique during exercise routines. By tracking the movement of the person's body and providing real-time feedback on their posture and form, the trainer can help users avoid common mistakes that can lead to injury or reduce the effectiveness of their workouts.
2. Personalized Coaching: Another benefit of using an AI fitness trainer is that it can provide personalized coaching based on the user's individual needs and goals. By analysing the user's movements and comparing them to ideal movements, the trainer can provide specific feedback and recommendations that are tailored to the user's needs. This can help users achieve their fitness goals more quickly and efficiently.
3. Real-time Feedback: An AI fitness trainer that uses Mediapipe and OpenCV can provide real-time feedback on a person's movements during exercise. This means that users can adjust their form and technique in real-time based on the feedback they receive, rather than having to wait for feedback after the workout is over. This can help users make immediate improvements to their form and get more out of their workouts.
4. Reduced risk of Injury: Finally, an AI fitness trainer that uses Mediapipe and OpenCV can help reduce the risk of injury during exercise routines. By providing feedback on posture and form, the trainer can help users avoid common mistakes that can lead to strain, sprains, and other injuries.

Overall, an AI fitness trainer that uses Mediapipe and OpenCV can provide several benefits to users. By providing personalized coaching, real-time feedback, and helping to improve form and technique, users can achieve their fitness goals more efficiently and with reduced risk of injury.

Conclusion

In conclusion, an AI fitness trainer using Python and MediaPipe can provide a personalized and interactive workout experience for users. With the combination of Python\'s machine learning and data analysis capabilities and MediaPipe\'s computer vision technology, the AI trainer can track and analyze user movements and provide real-time feedback and adjustments. The use of AI technology in fitness training can help improve efficiency, accuracy, and motivation, leading to better results and a more enjoyable workout experience.

References

[1] \"AI-powered Fitness Training: A Review of the Literature\" by J.C. Ibanez, K. Hu, and Y.L. Chen [2] Dr. S. M. Patil, Vaishnavi D. Patil, Kanchan M. Sharma, Shraddha S. Chaudhari and Smita S Talekar, “Artificial Intelligence-based Personal Fitness Trainer”, International Journal of Advanced Research in Science, Communication and Technology (IJARSCT), Vol. 2, Issue 1, November 2022. [3] Ji, Haoran, Karungaru Stephen Githinji, and Terada Kenji. \"AI Fitness Coach at Home using Image Recognition.\", 2022. [4] Kashish Jain, Jignasha Jadav, Manashvini Yadav and Dr Yogita Mane, ‘AI FITNESS TRAINER’, International Journal of Emerging Technologies and Innovative Research(UGC and ISSN Approved), no. 63975, vol.9, Issue 4, page no.g380-g385, April 2022. [5] Chin, Ji-Hyoung, Chanwook Do, and Minjung Kim. \"How to Increase Sport Facility Users’ Intention to Use AI Fitness Services: Based on the Technology Adoption Model.\" International Journal of Environmental Research and Public Health 19, no. 21, pp. 14453, November 2022. [6] Sharma, Ayushi, Jyotsna Pathak, Muskan Prakash and J. N. Singh. \"Object Detection using OpenCV and Python.\" In 2021 3rd International Conference on Advances in Computing, Communication Control and Networking (ICAC3N), pp. 501-505. IEEE, 2021. [7] Kanase, Rahul Ravikant, Akash Narayan Kumavat, Rohit Datta Sinalkar, and Sakshi Somani. \"Pose Estimation and Correcting Exercise Posture.\" In ITM Web of Conferences, vol. 40, p. 03031. EDP Sciences, 2021. [8] Gourangi Taware , Rohit Agrawal , Pratik Dhende , Prathamesh Jondhalekar, Shailesh Hule, 2021, “AI-based Workout Assistant and Fitness guide”, IJERT, Vol 10, Issue 11, November 2021. [9] Mokmin, Nur Azlina Mohamed. \"The effectiveness of a personalized virtual fitness trainer in teaching physical education by applying the artificial intelligent algorithm.\" Sciences 8, no. 5, pp. 258-264, October 2020. [10] Kreiss, Sven, Lorenzo Bertoni and Alexandre Alahi. “PifPaf: Composite Fields for Human Pose Estimation.” IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 11969-11978, 2019. [11] Naveenkumar, Mahamkali, and Ayyasamy Vadivel. \"OpenCV for computer vision applications.\" In Proceedings of national conference on big data and cloud computing (NCBDC’15), pp. 52-56. 2015. [12] H. Lee, J. Kim, and J.Y. Kim \"A Study on the Feasibility of AI-based Personalized Physical Training\" Korean Journal of Artificial Intelligence 9-2, p 15-21, 2021. [13] A. Casilli, F.P. De Simone, and A. De Natale \"Machine Learning for Physical Activity Recognition”, 2021. [14] Farrokhi, Alireza, Reza Farahbakhsh, Javad Rezazadeh, and Roberto Minerva. \"Application of Internet of Things and artificial intelligence for smart fitness: A survey.\" Computer Networks 189, p.107859, 2021. [15] Novatchkov, Hristo, and Arnold Baca. \"Artificial intelligence in sports on the example of weight training.\" Journal of sports science & medicine 12, no. 1, p 27, 2013.

Copyright

Copyright © 2023 Neha D, Dr. S. K. Manju Bargavi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.