With the rapid development of the intelligent spray-painting technology, the research and application of machine vision in the spray-painting robot system has attracted extensive attention. Reasonable trajectory of the spray-painting can ensure uniform paint thickness and reduce film defects. The closed-loop spraying system can be formed by combining the spraying trajectory with quality detection. In view of this, the research of trajectory planning for the spray-painting robot and quality detection of the paint film based on machine vision is reviewed. Firstly, the challenges, opportunities and machine vision technology of the spray-painting system are introduced in the rapid development of modern product manufacturing. Then, the research results of trajectory planning of the spray-painting robot and quality detection of the paint film are reviewed based on machine vision technology. The trajectory planning methods of the spray-painting robot based on machine vision are analyzed and discussed. The methods include the 3D reconstruction of the workpiece to be sprayed, the automatic trajectory planning based on the point cloud data and the compensation based on the visual servoing. This paper focuses on the application and research status of machine vision in the quality detection of the paint film. From two aspects of data enhancement and model selection, the potential solutions are provided to improve the performance of the algorithms for the quality detection based on deep learning in different tasks. Finally, the research methods and ideas of trajectory planning for the spray-painting robot and quality detection of the paint film are prospected and summarized based on machine vision. A reference for the development of the spraying system in the direction of intelligence and flexibility is also provided.