All existing data sources used to reconstruct 3D building models have certain restrictions. An eye-catching alternative is IndoorOSM, one of the most popular examples of the newly evolved crowdsourced geodata. The potential power of this rich and simple-formatted data source has been proven by many researches. However, a fatal flaw of IndoorOSM is also pointed out, which is its accuracy. Another promising data source that has been looked into is 2D architectural floor plans. They are also commonly available and full of detailed indoor information. Due to the inconsistencies and ambiguities existing among real-life floor plans, previous researches all are established on different user cases. Althogh sharing a common pipeline, they differ from each other in every step, from the data structure, processing procedure to the 3D reconstruction method. The combination of these two data sources can be beneficial, because architectural floor plans can offer IndoorOSM better accuracy and extensive indoor information while IndoorOSM can provide a unified data structure and 3D reconstruction workflow for information extracted from floor plans. Based on a throughout review of the characters of real-life floor plans, a set of rules are proposed to redraw architectural floor plans from real life. These rules mainly focus on reorganizing information contained in floor plans, taking advantages of the layering and blocking functions supported by CAD application. The original geometry and graphical representation in the raw floor plans is reserved as much as possible. Redrawing is only required when unstandardized representation is encountered. Then, an automatic process is accordingly presented to extract desired information from the redrawn floor plans into an IndoorOSM database. Finally, highly detailed CityGML LoD4 models with interior structures can be generated using a method proposed by Dr. Marcus Goetz. The pipeline is tested with several floor plans from real life for 3D reconstruction. User feedback validated the feasibility and efficiency of the redrawing rules.