Generating fast and accurate geospatial data and three-dimensional (3D) properties is essential for supporting multipurpose land management services. This study presents the CAdastre and Spatial map adjustment with spatial Computation for Automatic builDing dEtection and 3D generation (CASCADE-3D), a graphical user interface (GUI) developed for the automated reconstruction of 3D models at Levels of Detail (LOD) 1 and 2. CASCADE-3D integrates advanced deep-learning frameworks—including the Segment Anything Model (SAM) AI and Dynamic Graph Convolutional Neural Network (DGCNN)—to perform building outline detection and point cloud classification, subsequently reconstructing 3D models using true orthoimage acquired from unmanned aerial vehicles and building height models (BHMs). Photogrammetric processing using Structure-from-Motion and Multi-View Stereo techniques is initially applied to generate dense point clouds, true orthophotos, and BHMs. Building outlines are then extracted using SAM AI, a promptable segmentation system capable of zero-shot generalization to unfamiliar objects and images without the need for additional training. The CASCADE-3D GUI provides interactive digitization, automatic regularization, and refinement of the segmentation mask based on its primary orientation. Each BHM, representing the building height, is derived by classifying original point clouds using the DGCNN algorithm. This process enables the extraction of ground points and building classes from a digital terrain model (DTM) and BHM. The GUI facilitates the automated reconstruction of 3D building models at both LOD1 and LOD2. Accurate reconstruction of complex LOD2 models requires precise extraction of roof structures that captures the geometric configuration and orientation of roofs in intricate architectural forms. To achieve this objective, roof structure detection techniques were applied using each building's aspect. The study utilized point clouds and orthophotos of 1215 buildings, encompassing diverse architectural forms (with different shapes and patterns) and land cover types, across several provinces in Indonesia. The CASCADE-3D GUI was evaluated for detecting building outlines, roof structures, and LOD1/2 reconstruction. The results indicate that the reconstructed 3D building's geometry yielded an RMSE of 0.36 m. CASCADE-3D used the resulting building outlines, roof structures, DTMs, and BHMs to automatically reconstruct LOD1 and LOD2 building models, which were subsequently exported in CityJSON format.

Synchronous interactions from different locations have become a globally accepted modus of interaction since the COVID-19 outbreak. For centuries, professional cadastral survey activities always required an interaction modus whereby surveyors, neighboring landowners, and local officers were present simultaneously. During the systematic adjudication and land registration project in Indonesia, multiple problems in the land information systems emerged, which, up to date, remain unsolved. These include the presence of plots of land without a related title, incorrect demarcations in the field, and the listing of titles without a connection to a land plot. We argue that these problems emerged due to ineffective survey workflows, which draw on inflexible process steps. This research assesses how and how much the use of augmented and virtual reality (AR/VR) technologies can make land registration services more effective and expand collaboration in a synchronous and at distant manner (the so-called same time, different place principle). The tested cadastral survey workflows include the procedure for a first land titling, the one for land subdivision, and the updating and maintenance of the cadastral database. These are common cases that could potentially benefit from integrated uses of augmented and virtual reality applications. Mixed reality technologies using VR glasses are also tested as tools, allowing individuals, surveyors, and government officers to work together synchronously from different places via a web mediation dashboard. The work aims at providing alternatives for safe interactions of field surveyors with decision-making groups in their endeavors to reach fast and effective collaborative decisions on boundaries.