"Novel Multi-Exercise Pose Estimation (MEPE) Dataset "

"A novel Multi-Exercise Pose Estimation (MEPE) Dataset designed for accurate human motion analysis through joint angle calculation. The dataset comprises annotated images and video sequences covering multiple physical exercises, captured under diverse environmental conditions and viewpoints. Key body joints are labeled using state-of-the-art pose estimation frameworks, enabling precise extraction of skeletal keypoints and subsequent joint angle computation. The MEPE dataset supports the development and evaluation of algorithms for posture correction, fitness monitoring, rehabilitation assessment, and activity recognition. Additionally, it includes metadata such as exercise type, frame-level annotations, and calculated joint angles to facilitate supervised learning and performance benchmarking. The dataset aims to bridge the gap between generic pose datasets and exercise-specific applications by providing a comprehensive resource tailored for multi-exercise analysis and biomechanical interpretation."