AI-Enabled Interior Space Documentation with Instance Segmentation and Feature Extraction

Abstract

The study presents a novel approach to precisely gathering geographic data with iPhones with installed LIDAR sensors. Through dedicated iOS applications, this strategy enables high-fidelity environment scanning by exploiting consumer technologies. The procedure starts with capturing detailed point cloud data to improve its accuracy and usefulness. Next, features are extracted, and instances are segmented. The generated 3D models can be easily integrated into different CAD programs when exported in STL format using the open 3D library. This combination supports applications ranging from environmental monitoring and virtual reality development to urban planning and design. The technique shows how professional-level spatial analysis may be accomplished using consumer-grade sensors, offering insight into various sectors.

Additionally, the accessibility and user-friendliness of iPhones render this strategy very beneficial. By making sophisticated spatial analysis accessible to both professionals and amateurs without the need for costly, specialist equipment, it democratizes access to advanced geographical data collection methods. This study highlights how commonplace technology may make a big difference in industries that need accurate geographical data by providing a practical and economical replacement for more time-consuming techniques. In conclusion, this technique demonstrates how well iPhones equipped with LIDAR sensors may be used for professional-level geographic data collecting and processing. This approach highlights the transformational power of consumer technology in spatial analysis by emphasizing new possibilities for different applications by creating high-quality point cloud data and making it easier for it to be integrated into CAD software.