The PhD will explore the integration of artificial intelligence (AI) and conventional 3D surveying methods (photogrammetry and laser scanning) to enhance the precision, efficiency and completeness in architectural 3D documentation. In particular, the work will investigate how learning-based algorithms can automate and optimize various stages of the architectural 3D modeling process. Key objectives include, among others, the development of AI-driven tools for (i) generating complex geometries, (ii) improving completeness and efficiency in 3D documentation, (iii) modeling repetitive structures in building facade, (iv) complementing bottleneck of passive/active sensors. By leveraging AI, this research aims to revolutionize traditional architectural 3D documentation practices, offering innovative solutions that significantly reduce time and resource consumption while expanding the creative potential of the sector (including heritage). The findings of the PhD will contribute to the advancement of architectural 3D documentation, providing a framework for future research and practical applications.