The application of deep learning in stereo matching and disparity estimation: A bibliometric review

Estimating the depth of the 3D world from 2D images is a classic and important issue in computer vision, which has been widely studied for decades. With the remarkable effect of deep learning on various computer vision tasks, scholars have become increasingly interested in exploring stereo matching (i.e., disparity estimation) with deep learning. We reviewed related studies through bibliometrics, and especially extracted research hotspots and evolution context in the field, aiming to facilitate researchers in clarifying the positioning of their research and finding new inspiration. Specifically, we summarized the distribution of publication years, countries/regions, institutions, authors, research areas, document types, etc. of the research in this field. According to the analysis of information from these publications, we presented an overview of this field and divided its development into three stages: the preliminary exploration period (1999-2011), the gradual awakening period (2012-2016), and the vigorous development period (2017- present). Finally, we analyzed and predicted the future research directions.