Liquid composite molding (LCM) is recognized as one of the low-cost manufacturing technologies for composite materials, offering significant advantages for the production of large-size, complex, and high-strength composite components. However, defects such as porosity, dry spots, delamination, and warpage are prone to occur during the LCM process, which directly affect the mechanical properties, dimensional accuracy, and assembly quality of the components. As the demand for high-precision and low-cost manufacturing increases in the industry, real-time monitoring of the LCM process and defects has become a research hotspot. In this review, monitoring methods for key parameters such as the resin flow front during infiltration, as well as temperature, degree of cure, and residual stress during the curing process, were systematically surveyed from five aspects: fiber optics, electrical sensing, pressure, temperature, and ultrasonics. Detailed case studies were provided, and the applicability and limitations of each method were analyzed. Research progress in finite element simulation technology and advanced machine learning algorithms for LCM process and defect monitoring was reviewed, and the integration and engineering challenges of different methods were evaluated. Finally, the problems and future development trends in online monitoring systems for composite liquid molding processes were analyzed and summarized from the perspectives of multi-physics coupling, sensor network deployment, and real-time system integration.