The BIM research group from Tsinghua University’s School of Software recently won the Student Research Award at the 2025 buildingSMART International Summit in Berlin, Germany. Their project, BIMNet, offers an openBIM-based dataset and benchmark for scan-to-BIM processes.

The project was supervised by Professor Gu Ming from the School of Software and Assistant Researcher Gao Ge from the Beijing National Research Center for Information Science and Technology. The main contributors were students Liu Yudong, Huang Han, Ke Ziyi, Li Shengtao, and Zhang Xiaowei.

Award certificate (left) and the award ceremony

The buildingSMART open BIM Awards Program is an international BIM competition initiated and organized by buildingSMART, a globally recognized authority in the field of BIM.

In 2025, 129 projects worldwide were submitted for the award, among which 47 projects passed the preliminary selection. After rounds of international expert review and final presentations, BIMNet was the only project to win in the student research category.

The BIMNet dataset

Application ecosystem of BIMNet

A large-scale Scan-to-BIM dataset was developed, based on which BIMNet can reconstruct BIM models automatically from point clouds, encompassing both real-world point cloud data and manually modeled BIM data. It also proposes a multi-dimensional benchmark system and a standardized BIM workflow. This result provides the missing foundational support for Scan-to-BIM, facilitates AI model training, evaluation, and application in the architecture and engineering fields, and advances the intelligent and digital transformation of the Architecture, Engineering, and Construction (AEC) industry.

Editor: Li Han