Low-rank 3D mesh segmentation and labeling with structure guiding

• A low-rank representation model with structure guiding is designed to label 3D mesh.
• The tedious pre-training step existing in the data-driven approaches is eliminated.
• Just using a few examples, a test mesh can be labeled faithfully.
• Our method is robust to the seriously mislabeled examples.
• Correct labeling is got even if given examples include multiple object categories.

Semantic mesh segmentation and labeling is a fundamental problem in graphics. Conventional data-driven approaches usually employ a tedious offline pre-training process. Moreover, the number and especially the quality of the manually labeled examples challenge such strategies. In this paper, we develop a low-rank representation model with structure guiding to address these problems. The pre-training step is successfully eliminated and a test mesh can be labeled just using a few examples. As consistently labeling a large amount of meshes manually is a tedious procedure accompanied by inevitable mislabelings, our method is indeed more suitable for semantic mesh segmentation and labeling in real situations. In additional, by introducing the guiding from geometric similarity and labeling structure, and the robust ℓ2,1ℓ2,1 norm, our method generates correct labeling, even when the set of given examples contains multiple object categories or mislabeled meshes. Experimental results on the Princeton Segmentation Benchmark show that our approach outperforms the existing learning based methods.

Figure optionsDownload high-quality image (256 K)Download as PowerPoint slide