Team develops scan-to-BIM for reverse engineering from 3D vision data

Researchers astatine nan Korea Institute of Civil Engineering and Building Technology person developed building scan to BIM (Building Information Modeling)-based reverse engineering exertion required for 3D geospatial accusation modeling domains specified arsenic integer copy accusation modeling.

Scan to BIM-based reverse engineering exertion was studied to forestall errors and rework caused by manual activity erstwhile modeling shapes aliases BIM from 3D scan data. Existing reverse engineering activity is either done based connected a batch of manual activity aliases semi-automatically utilizing costly overseas software.

KICT Research Fellow Dr. Kang Tae-wook and an world associated investigation squad (from University of North Florida, Prof Jonghoon Kim; from Purdue University, Prof Kyubyung Kang; from IUPUI, Prof Dan Koo; and from The State University of New York, Prof Jongseong Brad Choi) developed Scan to BIM exertion customized for each reverse engineering intent based connected accumulated 3D vision, deep learning, and information processing pipeline technologies.

The Scan to BIM pipeline we are moving connected normalizes nan reverse engineerin engineering process into task units truthful they canㅇ beryllium assembled. Reverse engineering tasks are predefined, including LoD processing, classification, and geometry generation. This exertion provides Scan to BIM Description Language (SBDL) to link and combine tasks. Credit: Korea Institute of Civil Engineering and Building Technology(KICT)

This exertion automatically segments objects from 3D constituent unreality data, extracts style information, and creates BIM objects. Through this, accusation exemplary creation clip and rework required for integer twins tin beryllium dramatically reduced.

Productivity study cases were presented successful a paper published successful Applied Sciences. According to published data, this exertion improves reverse engineering productivity by 23.7 times and improves nan magnitude of modeling accusation by 110.21%. Additionally, this exertion tin beryllium customized depending connected nan purpose.

This R&D task believes that indoor and outdoor scan information tin beryllium utilized for installation attraction and operation, building management, infrastructure value inspection, and proptech plus digitization. Using Scan to BIM automation technology, we tin extract nan desired quantities from scan data, which is simply a elemental constituent data, and get nan dimensions of nan entity quickly. Credit: Korea Institute of Civil Engineering and Building Technology(KICT)

This exertion is utilized for 3D precision representation construction, 3D vectorizing, abnormal shape discovery during building tract information management, precise building management, and 3D digitalization of spatial accusation utilizing nan location, size, and guidance accusation of objects required from scanned image data and more. It tin beryllium utilized arsenic a guidelines exertion that tin beryllium included successful vision-based robotics and autonomous driving support systems.

The developed exertion is being tested on-site pinch a scan reverse engineering institution (BNG Co., Ltd) collaborating pinch Trimble Building Point successful Korea and overseas partner organizations. In addition, for nan improvement of nan industry, immoderate technologies were open-sourced and shared connected GitHub.

Dr. Kang said, “Currently, Scan to BIM reverse engineering automation exertion is difficult to specialize for each intent successful construction, and nan costs for nan AEC (architecture, engineering, and construction) manufacture to utilize nan exertion is besides high. The precocious developed Scan to BIM exertion will greatly thief AEC home business competitiveness.”