Multi-person 2D human pose estimation a benchmark for real-time applications

Abstract Human pose estimation (HPE) is a criticalcomponent of computer vision, enabling real-time ap-plications across various fields. This study benchmarkssix state-of-the-art frameworks — OpenPose, YOLO,RTMO, RTMPose, Sapiens, and MoveNet—in the context of 2D multi-person pose estimation in videos for real-time applications. Using the Panoptic dataset and standardized metrics, including Object Keypoint Similarity (OKS), Average Precision (AP), and Average Recall (AR), we evaluate their accuracy, speed, and hardware efficiency under realistic conditions such as occlusion and crowded scenes. Our analysis highlights the strengths and limitations of each framework, providing valuable insights to help practitioners select and deploy reliable HPE solutions in real-world applications. Mathematics Subject Classification (2020) MSC code1 · MSC code2 · more