Abstract: Traditional model-driven dim infrared small-target tracking algorithms find it difficult to balance the accuracy and real-time performance in complex scenes. After discussing the difficulties encountered during the spatial and temporal tracking of dim infrared small targets, this paper uses their reported tracking performances to classify and review tracking algorithms based on the data-driven use of spatiotemporal domain information. The basic strategies used by different neural network structures to analyze spatiotemporal domain information are reviewed, and the neural-network application scenarios in dim infrared small target tracking are analyzed. Finally, the development trends of dim infrared small target tracking research are discussed.