FICS 2024 Registrations are now open! Register Now
Idea - FICS

myTrainer

Team and Contact Details

Student Name School Degree Year Email
Rohan ManzoorSEECSMastersSecond[email protected]
Syed Zain AbbasSEECSMastersSecond[email protected]
Asad KhanSEECSPhDSecond[email protected]

Inter School Idea ? No
Do you need expertises from another area: No
If Yes please provide details of expertises you need:

Idea Details

Idea Name: myTrainer
Slogan: Your trainer in your pocket
Supervisor Name: Dr. Numan Khurshid
Supervisor Designation: Assistant Professor
Supervisor School: SEECS
Supervisor Department: SEECS
Contact number: +923339211849
Email ID: [email protected]
Abstract:
    One of the biggest hurdles in quality exercises is bad posture which can not only thwart progress, but can also cause injuries. Our app will analyze the posture of the person and identify the inaccuracies in it in realtime and then suggest them how to correct their pose so faster and safer workouts. Our platform will essentially be their own personal trainer helping them throughout the journey.
What is the unmet need in society that your idea will fulfill ?
    Gym trainers are expensive and not always available and there are no solutions that work on a mobile phone locally. We hope to provide a trainer to everyone and everywhere. Moreover our platform should work with existing platforms so other developers can incorporate it in their platforms aswell.
Who needs it ? How many would benefit ?
   Our target audiences are gyms and people interested in home fitness. Instead of having 2-3 trainers, Gyms will now able to provide a trainer to everyone. People doing fitness at home will now have their own trainer with them on their phone that they can use with our app or with other services
How will the solution works
    A RNN/transformer model will be fed poses detected from pre-recorded exercise videos using movenet. This will essentially build a catalogue or an understanding of how a particular exercise is done. We will then compute the pose of the person using our app and compare the poses with the pre-recorded video. Both pose will be fed to a neural network that will compute the disparity between the correct pose and our app user. Once that is identified, we will highlight the angles needed to be fixed.
Who are your competitors ? How is your solution different
    Adidas HIIT Workouts, FitOn Workouts, and Nike Training Club. Our USP is that unlike these apps which are nothing but prerecorded videos, we provide realtime feedback so that the persons can accurately follow the exercise. We will even be able to suggest different exercises based user performance
Status: new
Entry Date & Time: 2022-12-15 (1220)