Matches in SemOpenAlex for { <https://semopenalex.org/work/W2220394410> ?p ?o ?g. }
- W2220394410 abstract "Semantic segmentation is the task of assigning a class-label to each pixel in an image. We propose a region-based semantic segmentation framework which handles both full and weak supervision, and addresses three common problems: (1) Objects occur at multiple scales and therefore we should use regions at multiple scales. However, these regions are overlapping which creates conflicting class predictions at the pixel-level. (2) Class frequencies are highly imbalanced in realistic datasets. (3) Each pixel can only be assigned to a single class, which creates competition between classes. We address all three problems with a joint calibration method which optimizes a multi-class loss defined over the final pixel-level output labeling, as opposed to simply region classification. Our method outperforms the state-of-the-art on the popular SIFT Flow [18] dataset in both the fully and weakly supervised setting by a considerably margin (+6% and +10%, respectively)." @default.
- W2220394410 created "2016-06-24" @default.
- W2220394410 creator A5001917215 @default.
- W2220394410 creator A5003940396 @default.
- W2220394410 creator A5020938231 @default.
- W2220394410 date "2015-07-06" @default.
- W2220394410 modified "2023-09-27" @default.
- W2220394410 title "Joint Calibration for Semantic Segmentation" @default.
- W2220394410 cites W1529410181 @default.
- W2220394410 cites W1542723449 @default.
- W2220394410 cites W1618905105 @default.
- W2220394410 cites W1897260080 @default.
- W2220394410 cites W1901229278 @default.
- W2220394410 cites W1903029394 @default.
- W2220394410 cites W1909234690 @default.
- W2220394410 cites W1927251054 @default.
- W2220394410 cites W1938976761 @default.
- W2220394410 cites W1945608308 @default.
- W2220394410 cites W1993433125 @default.
- W2220394410 cites W1999478155 @default.
- W2220394410 cites W2013777358 @default.
- W2220394410 cites W2017691720 @default.
- W2220394410 cites W2022508996 @default.
- W2220394410 cites W2026581312 @default.
- W2220394410 cites W2051458493 @default.
- W2220394410 cites W2052196182 @default.
- W2220394410 cites W2066606526 @default.
- W2220394410 cites W2074005006 @default.
- W2220394410 cites W2076874408 @default.
- W2220394410 cites W2088049833 @default.
- W2220394410 cites W2102492119 @default.
- W2220394410 cites W2102605133 @default.
- W2220394410 cites W2105340328 @default.
- W2220394410 cites W2117539524 @default.
- W2220394410 cites W2123126017 @default.
- W2220394410 cites W2154644822 @default.
- W2220394410 cites W2158305599 @default.
- W2220394410 cites W2221898772 @default.
- W2220394410 cites W2950612966 @default.
- W2220394410 cites W2951221812 @default.
- W2220394410 cites W2951277909 @default.
- W2220394410 cites W2962835968 @default.
- W2220394410 cites W2963563573 @default.
- W2220394410 cites W2964288706 @default.
- W2220394410 cites W78159342 @default.
- W2220394410 hasPublicationYear "2015" @default.
- W2220394410 type Work @default.
- W2220394410 sameAs 2220394410 @default.
- W2220394410 citedByCount "2" @default.
- W2220394410 countsByYear W22203944102017 @default.
- W2220394410 countsByYear W22203944102018 @default.
- W2220394410 crossrefType "posted-content" @default.
- W2220394410 hasAuthorship W2220394410A5001917215 @default.
- W2220394410 hasAuthorship W2220394410A5003940396 @default.
- W2220394410 hasAuthorship W2220394410A5020938231 @default.
- W2220394410 hasConcept C105795698 @default.
- W2220394410 hasConcept C119857082 @default.
- W2220394410 hasConcept C127413603 @default.
- W2220394410 hasConcept C153180895 @default.
- W2220394410 hasConcept C154945302 @default.
- W2220394410 hasConcept C160633673 @default.
- W2220394410 hasConcept C162324750 @default.
- W2220394410 hasConcept C165838908 @default.
- W2220394410 hasConcept C170154142 @default.
- W2220394410 hasConcept C18555067 @default.
- W2220394410 hasConcept C187736073 @default.
- W2220394410 hasConcept C2777212361 @default.
- W2220394410 hasConcept C2780451532 @default.
- W2220394410 hasConcept C33923547 @default.
- W2220394410 hasConcept C41008148 @default.
- W2220394410 hasConcept C774472 @default.
- W2220394410 hasConcept C89600930 @default.
- W2220394410 hasConceptScore W2220394410C105795698 @default.
- W2220394410 hasConceptScore W2220394410C119857082 @default.
- W2220394410 hasConceptScore W2220394410C127413603 @default.
- W2220394410 hasConceptScore W2220394410C153180895 @default.
- W2220394410 hasConceptScore W2220394410C154945302 @default.
- W2220394410 hasConceptScore W2220394410C160633673 @default.
- W2220394410 hasConceptScore W2220394410C162324750 @default.
- W2220394410 hasConceptScore W2220394410C165838908 @default.
- W2220394410 hasConceptScore W2220394410C170154142 @default.
- W2220394410 hasConceptScore W2220394410C18555067 @default.
- W2220394410 hasConceptScore W2220394410C187736073 @default.
- W2220394410 hasConceptScore W2220394410C2777212361 @default.
- W2220394410 hasConceptScore W2220394410C2780451532 @default.
- W2220394410 hasConceptScore W2220394410C33923547 @default.
- W2220394410 hasConceptScore W2220394410C41008148 @default.
- W2220394410 hasConceptScore W2220394410C774472 @default.
- W2220394410 hasConceptScore W2220394410C89600930 @default.
- W2220394410 hasLocation W22203944101 @default.
- W2220394410 hasOpenAccess W2220394410 @default.
- W2220394410 hasPrimaryLocation W22203944101 @default.
- W2220394410 hasRelatedWork W1167987884 @default.
- W2220394410 hasRelatedWork W1203757169 @default.
- W2220394410 hasRelatedWork W1966984456 @default.
- W2220394410 hasRelatedWork W2211130039 @default.
- W2220394410 hasRelatedWork W2212714315 @default.
- W2220394410 hasRelatedWork W2761129952 @default.
- W2220394410 hasRelatedWork W2795574923 @default.
- W2220394410 hasRelatedWork W2905219103 @default.