Literature omics: a new approach to systematic review about pose estimation

Background: Pose estimation has been developed and is frequently used in motion capture and other applications. But it is still unclear “what exactly drives the development of pose estimation”. And no summary of the pose estimate milestones has been provided. Objective: The purpose of this paper is to examine the evolution of the field of pose estimation, identify significant developments (in terms of time, person, publications, institutions, nations, and their internal connections), and identify significant turning points and milestones. In addition, this paper aims to investigate pose estimation research. Methods: This paper conducts analysis of related literatures using the method of literatureomics based on 4873 literatures related to pose estimation that were retrieved from the Web of Science Core Collection. Perform co-authorship and keyword co-occurrence analysis and visualization using VOSviewer, and identify emerging keywords and reference literature using CiteSpace. Additionally, analyze milestone events over time by combining them with thematic modeling discussions. Conclusions: This study conducted a bibliometric analysis of over 20 years of research on pose estimation, identifying the countries, institutions, authors, journals, and publications involved in this field. Pose estimation has grown in popularity thanks to Deep Learning and COCO datasets, and it will be extensively employed in the surgical sector.