Traditional fitness tracking methods often rely on manual input, which can be imprecise and lead to suboptimal results, impacting the user's overall fitness journey. Our aim is to explore and develop a state-of-the art pose estimation system using Computer Vision and Machine Learning, to be used in the field of fitness.

Our model covers the following features
Exercise Recognition: Manual entry of exercises is error-prone and can disrupt the flow of a workout.
Leveraging OpenCV and deep learning, this system offers automated exercise recognition. Users can seamlessly
perform a variety of exercises, such as Pushup, Bicep Curl, and Squat, without the need for manual input.
Rep and Set Counting: Inaccurate rep and set counting can hinder the effectiveness of workouts, leading to under-training or over-training. Using OpenCV, the system sets angle thresholds, enabling accurate counting of repetitions ensuring users follow their workout plans with precision.

A web application that would assist in maintaining dietary planning, exercising, calculating calories, monitoring calorie intake and a voice assisted ai chatbot helping them follow their trajectory towards their nutrition and physique.