--- _id: '3198' abstract: - lang: eng text: 'In this paper we present a new approach for establishing correspondences between sparse image features related by an unknown non-rigid mapping and corrupted by clutter and occlusion, such as points extracted from a pair of images containing a human figure in distinct poses. We formulate this matching task as an energy minimization problem by defining a complex objective function of the appearance and the spatial arrangement of the features. Optimization of this energy is an instance of graph matching, which is in general a NP-hard problem. We describe a novel graph matching optimization technique, which we refer to as dual decomposition (DD), and demonstrate on a variety of examples that this method outperforms existing graph matching algorithms. In the majority of our examples DD is able to find the global minimum within a minute. The ability to globally optimize the objective allows us to accurately learn the parameters of our matching model from training examples. We show on several matching tasks that our learned model yields results superior to those of state-of-the-art methods. ' alternative_title: - LNCS author: - first_name: Lorenzo full_name: Torresani, Lorenzo last_name: Torresani - first_name: Vladimir full_name: Vladimir Kolmogorov id: 3D50B0BA-F248-11E8-B48F-1D18A9856A87 last_name: Kolmogorov - first_name: Carsten full_name: Rother, Carsten last_name: Rother citation: ama: 'Torresani L, Kolmogorov V, Rother C. Feature correspondence via graph matching: Models and global optimization. In: Vol 5303. Springer; 2008:596-609. doi:10.1007/978-3-540-88688-4_44' apa: 'Torresani, L., Kolmogorov, V., & Rother, C. (2008). Feature correspondence via graph matching: Models and global optimization (Vol. 5303, pp. 596–609). Presented at the ECCV: European Conference on Computer Vision, Springer. https://doi.org/10.1007/978-3-540-88688-4_44' chicago: 'Torresani, Lorenzo, Vladimir Kolmogorov, and Carsten Rother. “Feature Correspondence via Graph Matching: Models and Global Optimization,” 5303:596–609. Springer, 2008. https://doi.org/10.1007/978-3-540-88688-4_44.' ieee: 'L. Torresani, V. Kolmogorov, and C. Rother, “Feature correspondence via graph matching: Models and global optimization,” presented at the ECCV: European Conference on Computer Vision, 2008, vol. 5303, pp. 596–609.' ista: 'Torresani L, Kolmogorov V, Rother C. 2008. Feature correspondence via graph matching: Models and global optimization. ECCV: European Conference on Computer Vision, LNCS, vol. 5303, 596–609.' mla: 'Torresani, Lorenzo, et al. Feature Correspondence via Graph Matching: Models and Global Optimization. Vol. 5303, Springer, 2008, pp. 596–609, doi:10.1007/978-3-540-88688-4_44.' short: L. Torresani, V. Kolmogorov, C. Rother, in:, Springer, 2008, pp. 596–609. conference: name: 'ECCV: European Conference on Computer Vision' date_created: 2018-12-11T12:01:58Z date_published: 2008-01-01T00:00:00Z date_updated: 2021-01-12T07:41:44Z day: '01' doi: 10.1007/978-3-540-88688-4_44 extern: 1 intvolume: ' 5303' main_file_link: - open_access: '0' url: http://research-srv.microsoft.com/pubs/70610/eccv08-MatchingMRF.pdf month: '01' page: 596 - 609 publication_status: published publisher: Springer publist_id: '3485' quality_controlled: 0 status: public title: 'Feature correspondence via graph matching: Models and global optimization' type: conference volume: 5303 year: '2008' ...