<p>Metal laser cladding is widely used in surface strengthening, remanufacturing, and direct forming of complex parts. Avoiding the generation of defects such as cracks, surface unevenness, and poor fusion and precisely controlling the forming quality are the important challenges faced by metal laser cladding. The molten pool is the primary carrier of laser cladding process; monitoring and controlling the molten pool are crucial for addressing these challenges. This paper reviews research on laser cladding monitoring and control, focusing on image acquisition, feature extraction, process-features-properties correlation, and molten pool control. It has been observed that the side-axis shooting method, when combined with additional laser filling and equipped with a dimmer and filter, can effectively avoid strong light interference and capture clear images of the molten pool. The molten pool features can be extracted by threshold segmentation or deep learning methods according to the characteristics of the images. The correlation between process parameters, features, and coating properties is primarily established by developing mathematical models through statistics or regression models via machine learning. Furthermore, utilizing PID or improved PID control algorithms enables closed-loop control of the molten pool and indirect control of the coating process. The existing research has deficiencies in the relationship between the changes of the molten pool features and forming defects and quality, the accuracy of defect judgment, and the precision of process regulation. Multi-information fusion for defect judgment and multi-objective collaborative adjustment of process parameters are important research directions for precisely controlling the forming quality of laser cladding.</p>