Joint shape segmentation with linear programming

Qixing Huang, Vladlen Koltun, Leonidas Guibas

Research output: Chapter in Book/Report/Conference proceedingConference contribution

154 Scopus citations


We present an approach to segmenting shapes in a heterogenous shape database. Our approach segments the shapes jointly, utilizing features from multiple shapes to improve the segmentation of each. The approach is entirely unsupervised and is based on an integer quadratic programming formulation of the joint segmentation problem. The program optimizes over possible segmentations of individual shapes as well as over possible correspondences between segments from multiple shapes. The integer quadratic program is solved via a linear programming relaxation, using a block coordinate descent procedure that makes the optimization feasible for large databases. We evaluate the presented approach on the Princeton segmentation benchmark and show that joint shape segmentation significantly outperforms single-shape segmentation techniques. © 2011 ACM.
Original languageEnglish (US)
Title of host publicationProceedings of the 2011 SIGGRAPH Asia Conference on - SA '11
PublisherAssociation for Computing Machinery (ACM)
ISBN (Print)9781450308076
StatePublished - 2011
Externally publishedYes

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: We are grateful to Mirela Ben-Chen, Siddhartha Chaudhuri, and Evangelos Kalogerakis for their comments on this paper. This work was supported in part by NSF grants 0808515 and 1011228, a Stanford-KAUST AEA grant, and a Stanford Graduate Fellowship.
This publication acknowledges KAUST support, but has no KAUST affiliated authors.


Dive into the research topics of 'Joint shape segmentation with linear programming'. Together they form a unique fingerprint.

Cite this