LIFTR: A Virtual Personal Fitness Trainer - Powered by Computer Vision & AI

preview_player
Показать описание
DEMO AT 5:08.

Fourth year software engineering capstone project at The University of Western Ontario. Built with Python, OpenCV, Kivy, SocketIO, and PyTorch.
Рекомендации по теме
Комментарии
Автор

I started learning to code last Feb. I only vaguely knew I wanted to work for or on some type of fitness projects that help athletes optimize their performance and I recently learned about and fell into the rabbit hole of sports analytics and machine learning in sports and fitness. Fascinating stuff!

This is exactly the type of project that I would love to work on. Keep working on it and refining it, you definitely have something here. 🔥

videogameavatar
Автор

Hell of an app guys, Thanks for our final year project :)

nitinshukla
Автор

I am blown away by this. Congratulations on completing this

bored.
Автор

Interesting app, but a little over-simplified.
For example: You need to take into account the position of the wrist and further depth information you won't have through video(e.g. elbow too far back).
Nevertheless I think it's a very good start! With some more advanced training instructions and ways to approximate depth, that could be really helpful.

lazpap
Автор

This is amazing man good work dude I'll subscribe your channel

kunalprasad
Автор

That's such a cool way to use pose estimation for debug mode, it'd be great if you could break down some of the code!

ashwinsingh
Автор

Good stuff how's the whole process going

ghostman
Автор

Sir osm project... U did amazing job.... 👏👏👏👏plz do explain the code on the next

poojayu
Автор

sir can you plz explain the full code or how we use it after downloading sourcecode

Dudhondicha_maharaja_official
Автор

Love! Curious if you’ve gone further and tested this out?

keowkg
Автор

Have you considered giving the feedback via audio prompts.. in real-time, instead of user going back to the phone to get feedback. Maybe use a text-to-speech model.

zerowit
Автор

How can I add red or green signal at the joint position of human body?

coderaman_tech
Автор

Has anyone finalized a fully deployed app that is made for this purpose? I've been desperately searching.

Arturostunes
Автор

I am a personal trainer. Is there a way i can use this for online clients and brand it under my name

juliogeorgiou
Автор

Hello, is there any chance to get your code, or just lines about skeleton restrictions, It would very help me in my engeenering project, because I have to create app for helping people to improve their skill on gym. And also i have a question, do you use just openCV or also using tenserflow to improv quality of project?

krzysztofwicki
Автор

Please share the source code I have to take some idea

hammadqazi
Автор

Did you deploy your trained model somewhere for predictions?

gauravbhojwani
Автор

Nice work ! May I have a link to your source code ?

erwintchayem