Spatial relationships guide semantic component extraction from Gaussian splatting of traditional Chinese timber architecture
Year: 2026
Type: preprint
Abstract: Three-dimensional Gaussian splatting enables photorealistic documentation of architectural heritage, but extracting semantically accurate components remains challenging. We propose a framework for traditional Chinese timber architecture that combines spatial relationship-guided annotation with semantic–appearance decoupled Gaussian extraction. A spatial relationship graph guides a multimodal large language model in distinguishing visually similar components. Semantic assignments and filtering weights are then learned while keeping the reconstructed scene parameters fixed. On a Qing-dynasty timber-frame building with 22 component categories, the framework increases image-level mean intersection over union from 69.03% to 85.72% and extracted-object mean intersection over union from 28.01% to 69.24% compared with Heritage-3DGS. Average isolated-component peak signal-to-noise ratio increases from 17.05 to 24.30 dB, while full-scene rendering quality is preserved. The framework provides independently renderable component categories for semantic access to photorealistic architectural heritage records.
Source: Research Square
Authors: Lei Wang(Conceptualization, Investigation, Methodology, Software, Visualization, Writing – original draft), Ming Huang(Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing), Weifeng Ma(Software, Validation, Visualization), Dong Li(Data curation, Investigation), Xingxing Xiao(Formal analysis, Validation)
Language: English
Subfield: Geology
Field: Earth and Planetary Sciences
Domain: Physical Sciences
Open Access status: green
Award: G8218299626
OpenAlex ID: W7235891726
