Method segments 3D models by geometric texture patterns
Segmenting Geometric Textures of 3D Meshes
GraphicsComputer Vision and Pattern Recognition
Summary
Recognizing texture details on 3D shapes is important for many tasks, but current tools mostly identify large parts rather than detailed textures. The authors introduce a way to break 3D models into texture-based regions using a special model fine-tuned for this purpose. They also create a new benchmark to test such segmentation and develop an interactive tool to edit models and find similar ones based on these textures. Their approach works better than existing methods at capturing texture details on 3D surfaces.
What this means in practice
- •For heritage preservation teams: Identify and separate detailed surface textures on 3D scans of cultural artifacts to assist in digital archiving and restoration planning.
- •For 3d printing specialists: Segment 3D models by surface texture to optimize printing parameters or material choices for detailed parts during production.
Authors
Weidan Xiong, Kaijie Qu, Shihao Feng, Shunqi Fan, Rongyu Xiong, Hui Huang
Abstract
Recognizing geometric texture patterns on 3D models is critical for applications such as heritage protection, model editing, and 3D printing. Existing 3D model segmentation methods predominantly focus on identifying high-level semantic parts, thereby failing to capture middle-level textures. While recent Vision-Language Models can perform image segmentation with appropriate text prompts, they still struggle to recognize 3D geometric textures. In this work, we present a method that segments 3D mesh models based on middle-level geometric textures. We further introduce a new geometric texture segmentation benchmark and fine-tune SAM3 accordingly. The segmented texture regions are then clustered based on DINO feature similarity and lifted from 2D to 3D. Additionally, we develop an interactive tool to enable both 3D model editing based on our segmentation results and model retrieval from the dataset based on geometric and texture similarity. Experiments demonstrate that our method significantly outperforms existing methods in segmenting geometric textures and can facilitate various applications.