{"date_published":"2013-12-01T00:00:00Z","external_id":{"arxiv":["1310.1771"]},"publication_status":"published","month":"12","doi":"10.1109/ICCV.2013.288","citation":{"apa":"Gridchyn, I., &#38; Kolmogorov, V. (2013). Potts model, parametric maxflow and k-submodular functions (pp. 2320–2327). Presented at the ICCV: International Conference on Computer Vision, Sydney, Australia: IEEE. <a href=\"https://doi.org/10.1109/ICCV.2013.288\">https://doi.org/10.1109/ICCV.2013.288</a>","short":"I. Gridchyn, V. Kolmogorov, in:, IEEE, 2013, pp. 2320–2327.","ista":"Gridchyn I, Kolmogorov V. 2013. Potts model, parametric maxflow and k-submodular functions. ICCV: International Conference on Computer Vision, 2320–2327.","mla":"Gridchyn, Igor, and Vladimir Kolmogorov. <i>Potts Model, Parametric Maxflow and k-Submodular Functions</i>. IEEE, 2013, pp. 2320–27, doi:<a href=\"https://doi.org/10.1109/ICCV.2013.288\">10.1109/ICCV.2013.288</a>.","chicago":"Gridchyn, Igor, and Vladimir Kolmogorov. “Potts Model, Parametric Maxflow and k-Submodular Functions,” 2320–27. IEEE, 2013. <a href=\"https://doi.org/10.1109/ICCV.2013.288\">https://doi.org/10.1109/ICCV.2013.288</a>.","ieee":"I. Gridchyn and V. Kolmogorov, “Potts model, parametric maxflow and k-submodular functions,” presented at the ICCV: International Conference on Computer Vision, Sydney, Australia, 2013, pp. 2320–2327.","ama":"Gridchyn I, Kolmogorov V. Potts model, parametric maxflow and k-submodular functions. In: IEEE; 2013:2320-2327. doi:<a href=\"https://doi.org/10.1109/ICCV.2013.288\">10.1109/ICCV.2013.288</a>"},"publist_id":"4668","quality_controlled":"1","oa_version":"Preprint","year":"2013","oa":1,"title":"Potts model, parametric maxflow and k-submodular functions","language":[{"iso":"eng"}],"date_updated":"2021-01-12T06:56:28Z","conference":{"start_date":"2013-12-01","location":"Sydney, Australia","end_date":"2013-12-08","name":"ICCV: International Conference on Computer Vision"},"date_created":"2018-12-11T11:56:43Z","publisher":"IEEE","author":[{"last_name":"Gridchyn","full_name":"Gridchyn, Igor","first_name":"Igor","id":"4B60654C-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Vladimir","full_name":"Kolmogorov, Vladimir","last_name":"Kolmogorov","id":"3D50B0BA-F248-11E8-B48F-1D18A9856A87"}],"type":"conference","department":[{"_id":"JoCs"},{"_id":"VlKo"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","page":"2320 - 2327","main_file_link":[{"url":"http://arxiv.org/abs/1310.1771","open_access":"1"}],"day":"01","_id":"2276","status":"public","abstract":[{"text":"The problem of minimizing the Potts energy function frequently occurs in computer vision applications. One way to tackle this NP-hard problem was proposed by Kovtun [19, 20]. It identifies a part of an optimal solution by running k maxflow computations, where k is the number of labels. The number of “labeled” pixels can be significant in some applications, e.g. 50-93% in our tests for stereo. We show how to reduce the runtime to O (log k) maxflow computations (or one parametric maxflow computation). Furthermore, the output of our algorithm allows to speed-up the subsequent alpha expansion for the unlabeled part, or can be used as it is for time-critical applications. To derive our technique, we generalize the algorithm of Felzenszwalb et al. [7] for Tree Metrics . We also show a connection to k-submodular functions from combinatorial optimization, and discuss k-submodular relaxations for general energy functions.","lang":"eng"}]}