Table of Contents

1. Project Title.......................................................................................................................6
2. Project Overview...............................................................................................................6
3. Goals and Objectives.........................................................................................................6
4. Problem Statement...........................................................................................................7
5. High-level System Components.........................................................................................8
6. Optional Functional Units..................................................................................................8
7. Scope................................................................................................................................9
8. Impact of Project/Research on the Society........................................................................9
9. Gantt Chart......................................................................................................................10
10. Hardware and Software Requirement.............................................................................11
11. Tools and Technologies...................................................................................................11
12. Expertise of Team Members.............................................................................................12
13. References.......................................................................................................................12
List of Figures

Figure 1 | High-level System Components ………………………………………………… 8

Figure 2 | Gantt Chart ………………………………………………………………………….…… 10
List of Tables

Table 1 | Tools and Technologies ………………………………………………………………… 11

Table 2 | Expertise of Team Members …………………………………………………………. 12
 FYP – I: Project Proposal Date:05-Mar-24

1. Project Title
Application for Personalized Gym Training with Real-time Exercise Detection

2. Project Overview
In today's society, physical health and fitness are becoming increasingly important. Many of the
fitness methods of the past lacked flexible and immediate feedback and also gave people little
choice in terms of personalization and guidance. In addition, due to the development of
important technologies, the expectation and demand for ideas and special solutions that
demonstrate the potential of technology to improve the quality of information are increasing.
Personal training issues often include high costs, scheduling issues, and limitations. This makes
affordable and personal fitness workouts difficult to access to many people. Smart GymMate
steps in to provide a powerful, accessible platform for personal training that transcends the
limitations of traditional systems. Recently, deep learning has greatly pushed the boundaries of
HAR(Human Activity Recognition) on mobile and devices [1]. This knowledge led us to create
Smart GymMate, an Android app that provides fitness training with Realtime tracking. With
extensive knowledge of fitness goals and research methods, Smart GymMate offers a variety of
carefully selected training plans. With a plan tailored to their goals, people can begin their
exercise routines with confidence, no matter what their fitness goals are (weight loss, muscle
mass, or overall improvement). The concepts of freedom and adaptability are at the heart of
the Smart GymMate concept. Users can manage their health because they can customize their
workouts, regardless of their experience or fitness level. The platform meets a variety of
accuracy and efficiency needs, from beginners looking to build a solid foundation to
experienced athletes demanding maximum performance. Personal and immediate feedback is
an important part of the Smart GymMate experience. By exploring the use of technology and
personal training, users can receive instant feedback on the body movement and exercise form
detection to ensure safety and efficiency. This human activity is recognized by a series of
observation on the subject and his actions [3].

3. Goals and Objectives

Goals: The main goal is to enable people especially the beginners to reach their fitness
objectives by providing them with proper customized , easy to use mobile application powered
by artificial intelligence technology. AI Technology will help in real time human activity

5
 FYP – I: Project Proposal Date:05-Mar-24

detection and analysis, that includes the checking of angles for each exercise , to make sure
that user is executing exercises correctly in order to reduce the chances of injury and improve
efficiency of workouts. By prioritizing correct form, reducing injury risk and improved workout
efficiency our app revolutionizes fitness journeys.

Objectives:

 Personalized Exercise Plans: Mobile app feature that will use smart AI algorithms to
create exercise plans based on the user's health goals, interests, and fitness level.
 Diet Plan Recommendations: Enables a unique in app feature that provides
personalized meal plans that include the user's health goals and nutritional
preferences, providing support throughout their entire life.
 Exercise Detection: Use AI technology and camera integration to see the user's exercise
pattern and angle, providing instant feedback and guidance to ensure you're doing it
right and reduce the risk of injuries.

4. Problem Statement
When new fitness enthusiasts set out on their adventures, they encounter many obstacles that
hinder their growth and endanger their physical health. Confusing and frustrating training
schedules, personalized fitness plans, and information about key healthy behaviors are
frequently lacking. Obstacles like the cost of gym memberships and the scarcity of qualified
personal trainers further make these difficulties worse. Furthermore, novices frequently do
exercises incorrectly, which raises the risk of sprains, strains, and overuse injuries, among other
problems. These wounds discourage people from pursuing their exercise goals in addition to
impeding development. As a result, inexperienced fitness enthusiasts find it difficult to keep up
consistency in their routines, which makes it more difficult for them to reach their objectives. It
is critical to address these challenges—especially the predominance of harmful practices—in
order to assist beginners in realizing their goals of fitness.

6
 FYP – I: Project Proposal Date:05-Mar-24

5. High-level System Components

Figure 1: High Level System Components

6. Optional Functional Units

If we have the time and resources, we would like to add more features to our app. Our
objective is to consistently enhance the user experience while offering useful features. Among
these possible attributes are:

• Music Integration: Create customized workout playlists by syncing your favorite music
streaming services, which will increase your motivation while working out.
• Gym Accessories Marketplace: Easily browse and buy a large range of high-quality,
conveniently sized exercise essentials from top brands within the app.

7
 FYP – I: Project Proposal Date:05-Mar-24

7. Scope
The Smart GymMate Application for Personalized Gym Training with Real-time Exercise
Detection aims to revolutionize fitness training by integrating cutting-edge artificial intelligence
(AI) technology into a user-friendly mobile platform.. Based on user goals and fitness levels, it
provides customized meal plans and exercise regimens. It offers real-time workout detection
with instantaneous feedback via deep learning. Future updates will expand capabilities, with
the initial release concentrating on typical activities. Users have access to schedule exercises,
log nutrition, measure progress, set objectives, and access instructional content for correct
technique execution.

8. Impact of Project/Research on the Society

The Android application Smart GymMate for Personalized Gym Training with Real-time Exercise
Detection has the potential to significantly impact society by increasing the accessibility of
personalized fitness tools for all. With the app, users of all fitness levels can confidently and
successfully pursue their wellness and health goals by getting individualized training plans and
real-time activity feedback. Accessibility can help users attain improved physical health
outcomes by removing obstacles for those who were previously scared or hesitant to start a
fitness regimen. Monitoring human activities of daily living is an essential way of describing the
functional and health status of a human [2]. Furthermore, by encouraging good exercise
technique and reducing the risk of injury, Smart GymMate supports long-term health benefits
and promotes physical activity. Furthermore, by reaching a wider audience and encouraging a
wellness-conscious society, the app's emphasis on user-friendly interfaces and basic
functionality enhances inclusion. Generally speaking, Smart GymMate's impact on society
stems from its ability to inspire and support individuals on their journey towards improved
fitness and health, which eventually produces a population that is more active and healthier.

8
FYP – I: Project Proposal Date:05-Mar-24

9. Gantt Chart

Figure 2: Gantt Chart

9
 FYP – I: Project Proposal Date:05-Mar-24

10.Hardware and Software Requirement

Hardware Requirements:
 Processor: Intel(R) Core(TM) i5-6300U CPU
 RAM: 8 GB or higher
 Mobile Device(Good Camera Quality)
Software Requirements:

 Operating System: Android OS, iOS

 Android Studio

 Jupyter NoteBook.

 Firebase
 Adobe Figma (Logical designing of UI/UX)

11.Tools and Technologies

Flutter It Is a software tool for development of application.

OpenCV [4] A computer vision library offering tools for image and video
processing, crucial for analyzing and detecting exercises in real-
time

TensorFlow [5] An open-source machine learning framework used for building

and training machine learning models, enabling real-time
exercise detection.

Adobe Figma: A collaborative UI/UX design tool for creating visually appealing
and user-friendly interfaces for mobile applications.

10
 FYP – I: Project Proposal Date:05-Mar-24

12.Expertise of Team Members

Name Skills

Muhammad Arslan Flutter, ML, Python, Firebase, Back End Development

Shahzaib Hassan Android Studio, Adobe Figma, Front End Development
Shahzeb Asghar Flutter, ML, Firebase, Back End Development

13.References

[1] Zhang S, Li Y, Zhang S, Shahabi F, Xia S, Deng Y, Alshurafa N. Deep Learning in Human
Activity Recognition with Wearable Sensors: A Review on Advances. Sensors (Basel). 2022 Feb
14;22(4):1476. doi: 10.3390/s22041476. PMID: 35214377; PMCID: PMC8879042.
https://www.mdpi.com/1424-8220/22/4/1476
[2] A. Jalal, M. Mahmood and A. S. Hasan, "Multi-features descriptors for Human Activity
Tracking and Recognition in Indoor-Outdoor Environments," 2019 16th International Bhurban
Conference on Applied Sciences and Technology (IBCAST), Islamabad, Pakistan, 2019, pp. 371-
376, doi: 10.1109/IBCAST.2019.8667145.https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC8157856/
[3] Zhang, Shugang & Wei, Zhiqiang & Nie, Jie & Huang, Lei & Wang, Shuang & Li, Zhen. (2017).
A Review on Human Activity Recognition Using Vision-Based Method. Journal of Healthcare
Engineering.2017.131.10.1155/2017/3090343.https://www.hindawi.com/journals/jhe/
2017/3090343/
[4] [Online]. Available: https://opencv.org/
[5] [Online].Available https://www.tensorflow.org/

11