This paper makes a detailed introduction to the measurement principles of the optical surface absolute de? tection methods such as three?plane (spherical) mutual detection method, bispherical method, rotating average method, translational difference method (pseudo?shear interference method), parity function method, random ball method and rota? tion?translational method. It also describes the development status of absolute detection methods at home and abroad, and discusses the application fields of each absolute detection method and the corresponding technical limitations by compari? son. Then it makes prospects to the future trends of absolute detection technology from both physical implementation and algorithmic perspectives and proposes to increase the absolute detection accuracy of surface shape by improving the accu? racy of external mechanical structure. Finally, it analyzes the advantages of deep neural network algorithm and computa? tional optical imaging technology in the absolute detection process, and proposes that the absolute detection accuracy of optical surface shape can be further improved by combining the two technologies with the absolute detection methods. It can provide a useful reference for the research of absolute detection.