Scholarlink

Tibetan architecture is a unique style influenced by local natural and cultural factors, with significant aesthetic and practical value. Advances in technology, particularly 3D scanning, have made 3D point cloud models crucial for digitally preserving Tibetan architecture. Analyzing these models, especially feature extraction, is vital for understanding and preservation. Traditional methods often overlook global structures, focusing on local features. This study introduces a new feature descriptor, Enhancement Scaling Invariant Kernel Signature (ESIKS), combining BSIHKS and SIHWKS to extract geometric and topological information effectively. ESIKS improves feature extraction, model invariance, and precise segmentation of 3D point cloud models. Experiments show ESIKS excels in maintaining model invariance and provides robust theoretical and practical support for digital preservation and restoration of Tibetan architecture.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Feature Extraction and Analysis of 3D Ancient Building Point Cloud Model Based on Spectral Analysis: Taking Qutan Temple as an Example

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Jianpeng Zhang,
  • Dan Zhang,
  • Ning Wang,
  • Chenhao Xu,
  • Meihua Song,
  • Quanhong Peng

摘要

Tibetan architecture is a unique style influenced by local natural and cultural factors, with significant aesthetic and practical value. Advances in technology, particularly 3D scanning, have made 3D point cloud models crucial for digitally preserving Tibetan architecture. Analyzing these models, especially feature extraction, is vital for understanding and preservation. Traditional methods often overlook global structures, focusing on local features. This study introduces a new feature descriptor, Enhancement Scaling Invariant Kernel Signature (ESIKS), combining BSIHKS and SIHWKS to extract geometric and topological information effectively. ESIKS improves feature extraction, model invariance, and precise segmentation of 3D point cloud models. Experiments show ESIKS excels in maintaining model invariance and provides robust theoretical and practical support for digital preservation and restoration of Tibetan architecture.