In particle image velocimetry (PIV), neural network-based methods often face challenges when handling high-speed or complex nonlinear flows. These challenges include rapid changes in particle positions, which lead to difficulties in tracking and matching, limited feature scale extraction, and insufficient ability to capture effective features. To address these issues, a novel flow field estimation and dynamic particle tracking enhancement model LiteFlowNet-CL is proposed, based on the combination of ConvLSTM and the LiteFlowNet architecture. The study firstly enhances the ability of the LiteFlowNet model to identify and represent complex flow patterns, and then leverages the temporal modeling advantages of the ConvLSTM network to effectively suppress tracking errors of high-speed moving particles across different time steps, thereby significantly reducing the likelihood of particle image feature tracking loss. To validate the effectiveness of the proposed model, this paper conducted comparative performance tests and ablation experiments by using simulated particle images. Experimental results show that the improved velocity field estimation model achieved a root mean square error of 0.100 4. Compared with the classical LiteFlowNet optical flow estimation model, the error was reduced by 10.52%, while a further error reduction of 1.463% was observed when benchmarked against the widely adopted high-performance LiteFlowNet-en model in PIV domain. The proposed model was verified to effectively enhance the capability of capturing complex flow field characteristics in particle image velocimetry, with its error precision meeting experimental requirements for turbulence analysis. This achievement was recognized as providing a new technical pathway for PIV algorithm optimization, and its application value was confirmed in promoting the development of fluid mechanics experimental measurement technologies toward higher spatiotemporal resolution. The research methodology and implementation process were systematically described, with comprehensive quantitative comparisons presented to validate the performance improvements.