The advent of three-dimensional building footprint extraction and visualization has recently been explored due to its applications in urban planning, transportation, environmental monitoring, and modelling. Through the use of LiDAR (Light Detection and Ranging) data products such as Digital Terrain Model (DTM) and Digital Surface Model (DSM), useful information such as elevation, size, and shape can be obtained from the processed point clouds. Different studies regarding building extraction procedures have been conducted and explored. Algorithms have also been developed using LiDAR and GIS-based software that assume a rectangular form of polygons as building structures. As such, this study discusses an approach in developing a semiautomated building footprint extraction and three-dimensional visualization of the University of the Philippines (UP) main campus using solely LiDAR data and GIS-based open source software - QGIS Chugiak Version and GRASS 7.0. The UP-Diliman Campus situated in Quezon City, Philippines consists of various polygonal structured buildings which makes it a good subject area for this study. The proposed scheme composed of three major parts: LiDAR pre-processing, building footprint extraction and delineation, and three-dimensional building visualization. Normalized digital surface model (nDSM) can be generated using DTM and DSM. Under building footprint extraction and delineation, four parameters were set: height threshold, area scope, topographic modelling, and smoothing tolerance. For the classification by height, a threshold is set to remove structures with height less than the predetermined value of elevation. Classification by area removes the objects with areas that do not fall within the predetermined area scope. Topographic modelling is used mainly to separate building from other entities. Smoothing tolerance simplifies building outlines. The accuracy assessment for the extraction and delineation is quantified using ground truth data in comparison with the extracted polygons. As compared to manual digitization of building polygons, this semi-automation can be more efficient in extracting and delineating building footprint in areas with large scope. Experimental results indicate that the proposed scheme provides a promising solution for 3D building extraction and delineation using LiDAR data processed in QGIS Chugiak Version and GRASS 7.0. This methodology is a valuable tool for urban planning and modelling.