Published Paper Details:

Real-time body pose estimation stands as a pivotal component in computer vision, finding applications across an array of domains. This study delves into the amalgamation of OpenCV and MediaPipe, two robust libraries, to accomplish precise and efficient human body pose estimation in real-time. OpenCV, renowned for its computer vision functionalities, joins forces with MediaPipe, which furnishes pre-trained machine learning models explicitly crafted for keypoint estimation. This collaboration enables the accurate detection and continual tracking of human body landmarks.The methodology of this study centers on harnessing OpenCV's capabilities for managing video input and employing MediaPipe's pose estimation models for the identification of anatomical keypoints. OpenCV takes charge of vital video stream manipulations such as frame resizing, color space conversions, and noise reduction, optimizing the input data for MediaPipe's specialized models. Subsequently, MediaPipe adeptly pinpoints and tracks key body joints, empowering the real-time estimation of intricate human poses within live video streams or camera feeds. A comprehensive evaluation of this system encompasses scrutiny of its accuracy, real-time performance, and robustness under diverse conditions, encompassing scenarios of occlusion and varying environmental settings. The system's efficacy in detecting and persistently tracking keypoints, coupled with its real-time capabilities, unveils its potential in multifaceted applications such as sports analytics, healthcare, human-computer interaction, and beyond. The fusion of OpenCV and MediaPipe encapsulates a promising trajectory for real-time body pose estimation, laying a sturdy framework for precise human pose analysis. The study's findings contribute to propelling advancements in the realm of computer vision by furnishing a dependable and efficient solution for real-time pose estimation. These advancements hold the promise of impacting various industries and domains, hinting at significant strides in real-time pose estimation technology.