Construction method of multi-source Dougong point cloud dataset

Abstract

Abstract Dougong, a distinctive component of ancient wooden architecture, holds significant importance for the preservation and restoration of such structures. In the realm of cultural heritage preservation, the application of deep learning has gradually expanded, demonstrating remarkable effectiveness. Point cloud serving as a crucial source for Dougong, encapsulates various information, enabling support for tasks like Dougong point cloud classification and completion. The quality of Dougong datasets directly impacts the outcomes of deep learning, as they serve as the foundational data support for these tasks. However, due to the inherent characteristics of Dougong, such as coplanarity and occlusion, acquiring point cloud data is challenging, resulting in poor data quality and organizational difficulties. To address this, our study employs three data acquisition methods—real scanning, photo-generated point clouds, and model-sampled point clouds—to substantially augment the Dougong point cloud dataset. Further, through data augmentation, we enhance the dataset's volume and generalize its characteristics. This effort culminates in the creation of the Dougong Point Cloud Dataset (DG Dataset), poised to support deep learning tasks related to Dougong scenarios.