[{"project":[{"grant_number":"616160","name":"Discrete Optimization in Computer Vision: Theory and Practice","call_identifier":"FP7","_id":"25FBA906-B435-11E9-9278-68D0E5697425"}],"oa_version":"Submitted Version","month":"07","has_accepted_license":"1","conference":{"start_date":"2017-07-21","name":"CVPR: Computer Vision and Pattern Recognition","end_date":"2017-07-26","location":"Honolulu, HA, United States"},"language":[{"iso":"eng"}],"publication_identifier":{"isbn":["978-153860457-1"]},"publist_id":"6526","oa":1,"type":"conference","date_published":"2017-07-01T00:00:00Z","file":[{"date_created":"2019-01-18T12:52:46Z","checksum":"7e51dacefa693574581a32da3eff63dc","file_size":883264,"date_updated":"2020-07-14T12:48:15Z","content_type":"application/pdf","file_name":"Swoboda_A_Message_Passing_CVPR_2017_paper.pdf","relation":"main_file","access_level":"open_access","file_id":"5849","creator":"dernst"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","department":[{"_id":"VlKo"}],"date_created":"2018-12-11T11:49:11Z","article_processing_charge":"No","publication_status":"published","intvolume":"      2017","title":"A message passing algorithm for the minimum cost multicut problem","scopus_import":"1","_id":"915","author":[{"last_name":"Swoboda","first_name":"Paul","full_name":"Swoboda, Paul","id":"446560C6-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Andres","first_name":"Bjoern","full_name":"Andres, Bjoern"}],"publisher":"IEEE","quality_controlled":"1","ec_funded":1,"page":"4990-4999","file_date_updated":"2020-07-14T12:48:15Z","day":"01","doi":"10.1109/CVPR.2017.530","abstract":[{"lang":"eng","text":"We propose a dual decomposition and linear program relaxation of the NP-hard minimum cost multicut problem. Unlike other polyhedral relaxations of the multicut polytope, it is amenable to efficient optimization by message passing. Like other polyhedral relaxations, it can be tightened efficiently by cutting planes.  We define an algorithm that alternates between message passing and efficient separation of cycle- and odd-wheel inequalities. This algorithm is more efficient than state-of-the-art algorithms based on linear programming, including algorithms written in the framework of leading commercial software, as we show in experiments with large instances of the problem from applications in computer vision, biomedical image analysis and data mining."}],"year":"2017","citation":{"apa":"Swoboda, P., &#38; Andres, B. (2017). A message passing algorithm for the minimum cost multicut problem (Vol. 2017, pp. 4990–4999). Presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States: IEEE. <a href=\"https://doi.org/10.1109/CVPR.2017.530\">https://doi.org/10.1109/CVPR.2017.530</a>","ama":"Swoboda P, Andres B. A message passing algorithm for the minimum cost multicut problem. In: Vol 2017. IEEE; 2017:4990-4999. doi:<a href=\"https://doi.org/10.1109/CVPR.2017.530\">10.1109/CVPR.2017.530</a>","chicago":"Swoboda, Paul, and Bjoern Andres. “A Message Passing Algorithm for the Minimum Cost Multicut Problem,” 2017:4990–99. IEEE, 2017. <a href=\"https://doi.org/10.1109/CVPR.2017.530\">https://doi.org/10.1109/CVPR.2017.530</a>.","ieee":"P. Swoboda and B. Andres, “A message passing algorithm for the minimum cost multicut problem,” presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States, 2017, vol. 2017, pp. 4990–4999.","mla":"Swoboda, Paul, and Bjoern Andres. <i>A Message Passing Algorithm for the Minimum Cost Multicut Problem</i>. Vol. 2017, IEEE, 2017, pp. 4990–99, doi:<a href=\"https://doi.org/10.1109/CVPR.2017.530\">10.1109/CVPR.2017.530</a>.","short":"P. Swoboda, B. Andres, in:, IEEE, 2017, pp. 4990–4999.","ista":"Swoboda P, Andres B. 2017. A message passing algorithm for the minimum cost multicut problem. CVPR: Computer Vision and Pattern Recognition vol. 2017, 4990–4999."},"date_updated":"2023-09-26T15:43:27Z","external_id":{"isi":["000418371405009"]},"isi":1,"volume":2017,"ddc":["000"]},{"file":[{"access_level":"open_access","relation":"main_file","creator":"dernst","file_id":"5848","checksum":"e38a2740daad1ea178465843b5072906","file_size":944332,"date_created":"2019-01-18T12:49:38Z","content_type":"application/pdf","file_name":"2017_CVPR_Swoboda2.pdf","date_updated":"2020-07-14T12:48:15Z"}],"user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","status":"public","type":"conference","date_published":"2017-01-01T00:00:00Z","publication_identifier":{"isbn":["978-153860457-1"]},"oa":1,"publist_id":"6525","language":[{"iso":"eng"}],"conference":{"name":"CVPR: Computer Vision and Pattern Recognition","start_date":"2017-07-21","location":"Honolulu, HA, United States","end_date":"2017-07-26"},"has_accepted_license":"1","project":[{"name":"Discrete Optimization in Computer Vision: Theory and Practice","grant_number":"616160","call_identifier":"FP7","_id":"25FBA906-B435-11E9-9278-68D0E5697425"}],"oa_version":"Submitted Version","month":"01","volume":2017,"ddc":["000"],"citation":{"ieee":"P. Swoboda, C. Rother, C. Abu Alhaija, D. Kainmueller, and B. Savchynskyy, “A study of lagrangean decompositions and dual ascent solvers for graph matching,” presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States, 2017, vol. 2017, pp. 7062–7071.","chicago":"Swoboda, Paul, Carsten Rother, Carsten Abu Alhaija, Dagmar Kainmueller, and Bogdan Savchynskyy. “A Study of Lagrangean Decompositions and Dual Ascent Solvers for Graph Matching,” 2017:7062–71. IEEE, 2017. <a href=\"https://doi.org/10.1109/CVPR.2017.747\">https://doi.org/10.1109/CVPR.2017.747</a>.","ama":"Swoboda P, Rother C, Abu Alhaija C, Kainmueller D, Savchynskyy B. A study of lagrangean decompositions and dual ascent solvers for graph matching. In: Vol 2017. IEEE; 2017:7062-7071. doi:<a href=\"https://doi.org/10.1109/CVPR.2017.747\">10.1109/CVPR.2017.747</a>","apa":"Swoboda, P., Rother, C., Abu Alhaija, C., Kainmueller, D., &#38; Savchynskyy, B. (2017). A study of lagrangean decompositions and dual ascent solvers for graph matching (Vol. 2017, pp. 7062–7071). Presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States: IEEE. <a href=\"https://doi.org/10.1109/CVPR.2017.747\">https://doi.org/10.1109/CVPR.2017.747</a>","ista":"Swoboda P, Rother C, Abu Alhaija C, Kainmueller D, Savchynskyy B. 2017. A study of lagrangean decompositions and dual ascent solvers for graph matching. CVPR: Computer Vision and Pattern Recognition vol. 2017, 7062–7071.","short":"P. Swoboda, C. Rother, C. Abu Alhaija, D. Kainmueller, B. Savchynskyy, in:, IEEE, 2017, pp. 7062–7071.","mla":"Swoboda, Paul, et al. <i>A Study of Lagrangean Decompositions and Dual Ascent Solvers for Graph Matching</i>. Vol. 2017, IEEE, 2017, pp. 7062–71, doi:<a href=\"https://doi.org/10.1109/CVPR.2017.747\">10.1109/CVPR.2017.747</a>."},"year":"2017","date_updated":"2023-09-26T15:41:40Z","external_id":{"isi":["000418371407018"]},"isi":1,"day":"01","doi":"10.1109/CVPR.2017.747","abstract":[{"lang":"eng","text":"We study the quadratic assignment problem, in computer vision also known as graph matching. Two leading solvers for this problem optimize the Lagrange decomposition duals with sub-gradient and dual ascent (also known as message passing) updates. We explore this direction further and propose several additional Lagrangean relaxations of the graph matching problem along with corresponding algorithms, which are all based on a common dual ascent framework. Our extensive empirical evaluation gives several theoretical insights and suggests a new state-of-the-art anytime solver for the considered problem. Our improvement over state-of-the-art is particularly visible on a new dataset with large-scale sparse problem instances containing more than 500 graph nodes each."}],"quality_controlled":"1","ec_funded":1,"page":"7062-7071","file_date_updated":"2020-07-14T12:48:15Z","publisher":"IEEE","scopus_import":"1","_id":"916","author":[{"full_name":"Swoboda, Paul","last_name":"Swoboda","first_name":"Paul","id":"446560C6-F248-11E8-B48F-1D18A9856A87"},{"first_name":"Carsten","last_name":"Rother","full_name":"Rother, Carsten"},{"first_name":"Carsten","last_name":"Abu Alhaija","full_name":"Abu Alhaija, Carsten"},{"first_name":"Dagmar","last_name":"Kainmueller","full_name":"Kainmueller, Dagmar"},{"full_name":"Savchynskyy, Bogdan","first_name":"Bogdan","last_name":"Savchynskyy"}],"date_created":"2018-12-11T11:49:11Z","article_processing_charge":"No","department":[{"_id":"VlKo"}],"publication_status":"published","intvolume":"      2017","title":"A study of lagrangean decompositions and dual ascent solvers for graph matching"},{"status":"public","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","file":[{"file_id":"5847","creator":"dernst","access_level":"open_access","relation":"main_file","date_updated":"2020-07-14T12:48:15Z","content_type":"application/pdf","file_name":"2017_CVPR_Swoboda.pdf","date_created":"2019-01-18T12:45:55Z","checksum":"72fd291046bd8e5717961bd68f6b6f03","file_size":898652}],"type":"conference","date_published":"2017-07-01T00:00:00Z","oa":1,"publist_id":"6524","publication_identifier":{"isbn":["978-153860457-1"]},"language":[{"iso":"eng"}],"conference":{"start_date":"2017-07-21","name":"CVPR: Computer Vision and Pattern Recognition","end_date":"2017-07-26","location":"Honolulu, HA, United States"},"has_accepted_license":"1","month":"07","project":[{"grant_number":"616160","name":"Discrete Optimization in Computer Vision: Theory and Practice","call_identifier":"FP7","_id":"25FBA906-B435-11E9-9278-68D0E5697425"}],"oa_version":"Submitted Version","ddc":["000"],"volume":2017,"external_id":{"isi":["000418371405005"]},"isi":1,"year":"2017","citation":{"apa":"Swoboda, P., Kuske, J., &#38; Savchynskyy, B. (2017). A dual ascent framework for Lagrangean decomposition of combinatorial problems (Vol. 2017, pp. 4950–4960). Presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States: IEEE. <a href=\"https://doi.org/10.1109/CVPR.2017.526\">https://doi.org/10.1109/CVPR.2017.526</a>","ama":"Swoboda P, Kuske J, Savchynskyy B. A dual ascent framework for Lagrangean decomposition of combinatorial problems. In: Vol 2017. IEEE; 2017:4950-4960. doi:<a href=\"https://doi.org/10.1109/CVPR.2017.526\">10.1109/CVPR.2017.526</a>","chicago":"Swoboda, Paul, Jan Kuske, and Bogdan Savchynskyy. “A Dual Ascent Framework for Lagrangean Decomposition of Combinatorial Problems,” 2017:4950–60. IEEE, 2017. <a href=\"https://doi.org/10.1109/CVPR.2017.526\">https://doi.org/10.1109/CVPR.2017.526</a>.","ieee":"P. Swoboda, J. Kuske, and B. Savchynskyy, “A dual ascent framework for Lagrangean decomposition of combinatorial problems,” presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States, 2017, vol. 2017, pp. 4950–4960.","short":"P. Swoboda, J. Kuske, B. Savchynskyy, in:, IEEE, 2017, pp. 4950–4960.","mla":"Swoboda, Paul, et al. <i>A Dual Ascent Framework for Lagrangean Decomposition of Combinatorial Problems</i>. Vol. 2017, IEEE, 2017, pp. 4950–60, doi:<a href=\"https://doi.org/10.1109/CVPR.2017.526\">10.1109/CVPR.2017.526</a>.","ista":"Swoboda P, Kuske J, Savchynskyy B. 2017. A dual ascent framework for Lagrangean decomposition of combinatorial problems. CVPR: Computer Vision and Pattern Recognition vol. 2017, 4950–4960."},"date_updated":"2023-09-26T15:41:11Z","abstract":[{"text":"We  propose  a  general  dual  ascent  framework  for  Lagrangean decomposition of combinatorial problems.  Although methods of this type have shown their efficiency for a number of problems, so far there was no general algorithm applicable to multiple problem types. In this work, we propose such a general algorithm. It depends on several parameters, which can be used to optimize its performance in each particular setting. We demonstrate efficacy of our method on graph matching and multicut problems, where it outperforms state-of-the-art solvers including those based on subgradient optimization and off-the-shelf linear programming solvers.","lang":"eng"}],"day":"01","doi":"10.1109/CVPR.2017.526","file_date_updated":"2020-07-14T12:48:15Z","quality_controlled":"1","ec_funded":1,"page":"4950-4960","publisher":"IEEE","author":[{"first_name":"Paul","last_name":"Swoboda","full_name":"Swoboda, Paul","id":"446560C6-F248-11E8-B48F-1D18A9856A87"},{"last_name":"Kuske","first_name":"Jan","full_name":"Kuske, Jan"},{"full_name":"Savchynskyy, Bogdan","last_name":"Savchynskyy","first_name":"Bogdan"}],"scopus_import":"1","_id":"917","intvolume":"      2017","title":"A dual ascent framework for Lagrangean decomposition of combinatorial problems","date_created":"2018-12-11T11:49:11Z","department":[{"_id":"VlKo"}],"article_processing_charge":"No","publication_status":"published"},{"publisher":"IEEE","page":"5533 - 5542","ec_funded":1,"quality_controlled":"1","publication_status":"published","article_processing_charge":"No","date_created":"2018-12-11T11:49:37Z","department":[{"_id":"ChLa"},{"_id":"ChWo"}],"title":"iCaRL: Incremental classifier and representation learning","intvolume":"      2017","_id":"998","scopus_import":"1","author":[{"full_name":"Rebuffi, Sylvestre Alvise","first_name":"Sylvestre Alvise","last_name":"Rebuffi"},{"full_name":"Kolesnikov, Alexander","first_name":"Alexander","last_name":"Kolesnikov","id":"2D157DB6-F248-11E8-B48F-1D18A9856A87"},{"id":"4DD40360-F248-11E8-B48F-1D18A9856A87","last_name":"Sperl","first_name":"Georg","full_name":"Sperl, Georg"},{"id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","first_name":"Christoph","full_name":"Lampert, Christoph","orcid":"0000-0001-8622-7887"}],"volume":2017,"doi":"10.1109/CVPR.2017.587","day":"14","abstract":[{"lang":"eng","text":"A major open problem on the road to artificial intelligence is the development of incrementally learning systems that learn about more and more concepts over time from a stream of data. In this work, we introduce a new training strategy, iCaRL, that allows learning in such a class-incremental way: only the training data for a small number of classes has to be present at the same time and new classes can be added progressively. iCaRL learns strong classifiers and a data representation simultaneously. This distinguishes it from earlier works that were fundamentally limited to fixed data representations and therefore incompatible with deep learning architectures. We show by experiments on CIFAR-100 and ImageNet ILSVRC 2012 data that iCaRL can learn many classes incrementally over a long period of time where other strategies quickly fail. "}],"date_updated":"2023-09-22T09:51:58Z","year":"2017","citation":{"ieee":"S. A. Rebuffi, A. Kolesnikov, G. Sperl, and C. Lampert, “iCaRL: Incremental classifier and representation learning,” presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States, 2017, vol. 2017, pp. 5533–5542.","chicago":"Rebuffi, Sylvestre Alvise, Alexander Kolesnikov, Georg Sperl, and Christoph Lampert. “ICaRL: Incremental Classifier and Representation Learning,” 2017:5533–42. IEEE, 2017. <a href=\"https://doi.org/10.1109/CVPR.2017.587\">https://doi.org/10.1109/CVPR.2017.587</a>.","apa":"Rebuffi, S. A., Kolesnikov, A., Sperl, G., &#38; Lampert, C. (2017). iCaRL: Incremental classifier and representation learning (Vol. 2017, pp. 5533–5542). Presented at the CVPR: Computer Vision and Pattern Recognition, Honolulu, HA, United States: IEEE. <a href=\"https://doi.org/10.1109/CVPR.2017.587\">https://doi.org/10.1109/CVPR.2017.587</a>","ama":"Rebuffi SA, Kolesnikov A, Sperl G, Lampert C. iCaRL: Incremental classifier and representation learning. In: Vol 2017. IEEE; 2017:5533-5542. doi:<a href=\"https://doi.org/10.1109/CVPR.2017.587\">10.1109/CVPR.2017.587</a>","ista":"Rebuffi SA, Kolesnikov A, Sperl G, Lampert C. 2017. iCaRL: Incremental classifier and representation learning. CVPR: Computer Vision and Pattern Recognition vol. 2017, 5533–5542.","short":"S.A. Rebuffi, A. Kolesnikov, G. Sperl, C. Lampert, in:, IEEE, 2017, pp. 5533–5542.","mla":"Rebuffi, Sylvestre Alvise, et al. <i>ICaRL: Incremental Classifier and Representation Learning</i>. Vol. 2017, IEEE, 2017, pp. 5533–42, doi:<a href=\"https://doi.org/10.1109/CVPR.2017.587\">10.1109/CVPR.2017.587</a>."},"isi":1,"external_id":{"isi":["000418371405066"]},"conference":{"location":"Honolulu, HA, United States","end_date":"2017-07-26","start_date":"2017-07-21","name":"CVPR: Computer Vision and Pattern Recognition"},"language":[{"iso":"eng"}],"oa_version":"Submitted Version","project":[{"grant_number":"308036","name":"Lifelong Learning of Visual Scene Understanding","_id":"2532554C-B435-11E9-9278-68D0E5697425","call_identifier":"FP7"}],"month":"04","main_file_link":[{"open_access":"1","url":"https://arxiv.org/abs/1611.07725"}],"status":"public","user_id":"c635000d-4b10-11ee-a964-aac5a93f6ac1","publication_identifier":{"isbn":["978-153860457-1"]},"oa":1,"publist_id":"6400","date_published":"2017-04-14T00:00:00Z","type":"conference"}]