Disasters usually result in many casualties, and it is very difficult to avoid. Especially in Indonesia, where disaster (natural disasters) occurs frequently, great losses have been experienced. Authorities have worked to reduce losses during disasters by planning evacuation procedures. One of the important objectives is to save lives. Therefore, the capability to identify a victim's location during a disaster becomes crucial, particularly when it occurs indoors. Wi-Fi fingerprinting-based indoor positioning system (IPS) has become a widely used technique to predict the location of users indoors. The reason is that Wi-Fi infrastructure is widely deployed in indoor environments. Thus, Wi-Fi-based IPSs can be deployed instantly at a very low cost. However, Wi-Fi fingerprinting-based IPSs usually incur high overhead due to the requirement for manually collecting and labeling fingerprint data. A commonly used approach is gridbased fingerprint collection. The building is divided into several grids, and fingerprints are taken from each grid in sequence. Hence, the overhead increases with the building size. To eliminate this overhead, we propose automatic fingerprint collection and labeling by leveraging readily available information in a smart campus environment, such as student information, course information, and building map data. By exploiting this information, we can run seamless service on students' or employees' devices to collect fingerprint data and label the data according to the course enrollment and schedule or office room assignment. We have implemented the proposed IPS and made the source code publicly available. The performance evaluation shows that the proposed IPS has an accuracy as high as that of the IPS with manual fingerprint collection and labeling. Meanwhile, the proposed IPS successfully eliminates the labor burden of collecting and labeling fingerprint data. © 2025 IEEE.