Scholarlink

Cleft lips are one of the most common birth defects worldwide, and a quantitative method for analyzing facial symmetry is required to enable more effective surgical planning. To realize a detailed analysis of the 3D facial data, facial part segmentation from the 3D point cloud plays an important role, however this area has not been explored yet. In this paper, we introduce a new deep learning-based approach for facial part segmentation from 3D point cloud data. Our main contributions are as follows: (1) we present a new approach for creating a large-scale synthetic dataset of facial 3D point clouds, and (2) we test the performance of a deep learning model using the synthetic dataset and our private dataset. The experimental results demonstrate the great potential of deep learning for understanding 3D point cloud data of the face.

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

A Novel Deep Learning-Based Approach for Facial Part Segmentation from 3D Point Cloud Data

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Narumi Kihara,
  • Namiko Kimura-Nomoto,
  • Takako Okawachi,
  • Guangxu Li,
  • Norifumi Nakamura,
  • Tohru Kamiya

摘要

Cleft lips are one of the most common birth defects worldwide, and a quantitative method for analyzing facial symmetry is required to enable more effective surgical planning. To realize a detailed analysis of the 3D facial data, facial part segmentation from the 3D point cloud plays an important role, however this area has not been explored yet. In this paper, we introduce a new deep learning-based approach for facial part segmentation from 3D point cloud data. Our main contributions are as follows: (1) we present a new approach for creating a large-scale synthetic dataset of facial 3D point clouds, and (2) we test the performance of a deep learning model using the synthetic dataset and our private dataset. The experimental results demonstrate the great potential of deep learning for understanding 3D point cloud data of the face.