Yoga Pose Detection using Keras Neural Network

Abstract— Our project aims to create an intricate deep learning system that can accurately detect human poses. We plan on utilizing various technologies such as Mediapipe, OpenCV, TensorFlow, Keras, and NumPy to develop this model. The project focuses on detecting human body poses accurately, particularly for the purpose of yoga. The project begins with the implementation of the Mediapipe pose detection library to capture the human body pose. Next, a simple Dense network model is developed using Keras, which is trained on the data obtained from the Mediapipe pose detection library. The training process involves feeding the model with input data and corresponding output labels to optimize the model for better predictions. Once the model is trained, an inference file is created to use the model for predicting the human body pose.