In the field of computer vision, research on human pose estimation aims to understand human movements by analyzing individual camera images. However, identifying individuals in multi-people pose estimation remains a challenging issue. This study explores ID assignment technology to address the technical challenge of identifying individuals and maintaining consistent identities over consecutive images. Previous studies often required additional training to perform optimally in the wild environments. Therefore, in order to address the limitation of the previous studies, we propose a novel methodology that performs real-time image analysis and evaluation without the need for any prior training or additional datasets. By identifying the number of people and analyzing individual characteristics in each frame, we compare the characteristics between the current and previous frames, assigning the same ID to the individuals showing the highest similarity. This methodology enhances the applicability of human pose estimation systems in the wild and in real-time without training. We conducted experiments across four distinct scenarios and observed that the average ID assignment accuracy across all data was approximately 86.01%. Consequently, our research is expected to make significant contributions to various applications, including human-computer interaction, sports analysis, healthcare, and surveillance systems.