DC Field | Value | Language |
---|---|---|
dc.contributor.author | Ham, Bumsub | - |
dc.contributor.author | CHO, MINSU | - |
dc.contributor.author | Schmid, Cordelia | - |
dc.contributor.author | Ponce, Jean | - |
dc.date.accessioned | 2018-05-04T02:33:53Z | - |
dc.date.available | 2018-05-04T02:33:53Z | - |
dc.date.created | 2018-03-05 | - |
dc.date.issued | 2018-01 | - |
dc.identifier.issn | 0162-8828 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/41190 | - |
dc.description.abstract | Finding image correspondences remains a challenging problem in the presence of intra-class variations and large changes in scene layout. Semantic flow methods are designed to handle images depicting different instances of the same object or scene category. We introduce a novel approach to semantic flow, dubbed proposal flow, that establishes reliable correspondences using object proposals. Unlike prevailing semantic flow approaches that operate on pixels or regularly sampled local regions, proposal flow benefits from the characteristics of modern object proposals, that exhibit high repeatability at multiple scales, and can take advantage of both local and geometric consistency constraints among proposals. We also show that the corresponding sparse proposal flow can effectively be transformed into a conventional dense flow field. We introduce two new challenging datasets that can be used to evaluate both general semantic flow techniques and region-based approaches such as proposal flow. We use these benchmarks to compare different matching algorithms, object proposals, and region features within proposal flow, to the state of the art in semantic flow. This comparison, along with experiments on standard datasets, demonstrates that proposal flow significantly outperforms existing semantic flow methods in various settings. | - |
dc.language | English | - |
dc.publisher | IEEE | - |
dc.relation.isPartOf | IEEE Transactions on Pattern Analysis and Machine Intelligence | - |
dc.subject | Artificial intelligence | - |
dc.subject | Clutter (information theory) | - |
dc.subject | Computer vision | - |
dc.subject | Electronic mail | - |
dc.subject | Robustness (control systems) | - |
dc.subject | Benchmark testing | - |
dc.subject | dense scene correspondence | - |
dc.subject | Object proposals | - |
dc.subject | Optical imaging | - |
dc.subject | Proposals | - |
dc.subject | Semantics | - |
dc.title | Proposal Flow: Semantic Correspondences from Object Proposals | - |
dc.type | Article | - |
dc.identifier.doi | 10.1109/TPAMI.2017.2724510 | - |
dc.type.rims | ART | - |
dc.identifier.bibliographicCitation | IEEE Transactions on Pattern Analysis and Machine Intelligence, v.40, no.7, pp.1711 - 1725 | - |
dc.identifier.wosid | 000434294800013 | - |
dc.citation.endPage | 1725 | - |
dc.citation.number | 7 | - |
dc.citation.startPage | 1711 | - |
dc.citation.title | IEEE Transactions on Pattern Analysis and Machine Intelligence | - |
dc.citation.volume | 40 | - |
dc.contributor.affiliatedAuthor | CHO, MINSU | - |
dc.identifier.scopusid | 2-s2.0-85023646116 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
dc.type.docType | ARTICLE | - |
dc.subject.keywordPlus | BASE-LINE STEREO | - |
dc.subject.keywordPlus | DENSE CORRESPONDENCES | - |
dc.subject.keywordPlus | LOCATION | - |
dc.subject.keywordPlus | MODELS | - |
dc.subject.keywordPlus | PARTS | - |
dc.subject.keywordAuthor | Semantic flow | - |
dc.subject.keywordAuthor | object proposals | - |
dc.subject.keywordAuthor | scene alignment | - |
dc.subject.keywordAuthor | dense scene correspondence | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Artificial Intelligence | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.