In this project, The fusion of Computer vision and machine learning was used to analyze bicep curls, so that one can check the accuracy of a bicep curl in real time and provide insights into the bicep curl. The proposed setup efficiently examines input videos, recording and calculating the angles formed by the left and right arms in the video during bicep exercises using Opencv and Mediapipe features. Use of a Logistic Regression model expands the project's potential by enabling the prediction of bicep curl accuracy based on right and left arm data. This combination of computer vision and machine learning is a useful tool for fitness enthusiasts, offering an integrated approach to exercise monitoring and improvement. The study not only advances the understanding of bicep curl exercise but also paves the path for the use of automated assessments, contributing to the developing field of technology-assisted exercise analysis. This research not only examines blending computer vision and machine learning for bicep curls but also proposes an additional use of this technology for overall activity monitoring. The goal is to expand the scope beyond bicep curls to enhance the methods for predicting and sustaining exercise accuracy through a variety of exercises. This approach attempts to lay the groundwork for a comprehensive application-based system that maintains accuracy criteria across a wide range of exercises.