Matches in SemOpenAlex for { <https://semopenalex.org/work/W2895589541> ?p ?o ?g. }
- W2895589541 endingPage "4252" @default.
- W2895589541 startingPage "4243" @default.
- W2895589541 abstract "This paper focuses on weakly supervised image understanding, in which the semantic labels are available only at image-level, without the specific object or scene location in an image. Existing algorithms implicitly assume that image-level labels are error-free, which might be too restrictive. In practice, image labels obtained from the pretrained predictors are easily contaminated. To solve this problem, we propose a novel algorithm for weakly supervised segmentation when only noisy image labels are available during training. More specifically, a semantic space is constructed first by encoding image labels through a graphlet (i.e., superpixel cluster) embedding process. Then, we observe that in the semantic space, the distribution of graphlets from images with a same label remains stable, regardless of the noises in image labels. Therefore, we propose a generative model, called latent stability analysis, to discover the stable patterns from images with noisy labels. Inferring graphlet semantics by making use of these mid-level stable patterns is much more secure and accurate than directly transferring noisy image-level labels into different regions. Finally, we calculate the semantics of each superpixel using maximum majority voting of its correlated graphlets. Comprehensive experimental results show that our algorithm performs impressively when the image labels are predicted by either the hand-crafted or deeply learned image descriptors." @default.
- W2895589541 created "2018-10-12" @default.
- W2895589541 creator A5003583859 @default.
- W2895589541 creator A5031755593 @default.
- W2895589541 creator A5049254107 @default.
- W2895589541 creator A5072203134 @default.
- W2895589541 creator A5072690470 @default.
- W2895589541 creator A5073070877 @default.
- W2895589541 creator A5082634513 @default.
- W2895589541 date "2019-12-01" @default.
- W2895589541 modified "2023-10-15" @default.
- W2895589541 title "Learning Latent Stable Patterns for Image Understanding With Weak and Noisy Labels" @default.
- W2895589541 cites W1566135517 @default.
- W2895589541 cites W1971756996 @default.
- W2895589541 cites W1976977741 @default.
- W2895589541 cites W1983521647 @default.
- W2895589541 cites W1991598519 @default.
- W2895589541 cites W1993433125 @default.
- W2895589541 cites W1999478155 @default.
- W2895589541 cites W2023081159 @default.
- W2895589541 cites W2026581312 @default.
- W2895589541 cites W2029731618 @default.
- W2895589541 cites W2031489346 @default.
- W2895589541 cites W2040999325 @default.
- W2895589541 cites W2046382188 @default.
- W2895589541 cites W2049612634 @default.
- W2895589541 cites W2071901160 @default.
- W2895589541 cites W2073212499 @default.
- W2895589541 cites W2075760200 @default.
- W2895589541 cites W2076363162 @default.
- W2895589541 cites W2109154902 @default.
- W2895589541 cites W2115467367 @default.
- W2895589541 cites W2117553576 @default.
- W2895589541 cites W2118246710 @default.
- W2895589541 cites W2119531662 @default.
- W2895589541 cites W2121947440 @default.
- W2895589541 cites W2123023145 @default.
- W2895589541 cites W2125849446 @default.
- W2895589541 cites W2129947832 @default.
- W2895589541 cites W2139172286 @default.
- W2895589541 cites W2140539195 @default.
- W2895589541 cites W2145508099 @default.
- W2895589541 cites W2153125595 @default.
- W2895589541 cites W2158427031 @default.
- W2895589541 cites W2162915993 @default.
- W2895589541 cites W2535516436 @default.
- W2895589541 cites W4293052541 @default.
- W2895589541 doi "https://doi.org/10.1109/tcyb.2018.2861419" @default.
- W2895589541 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30296245" @default.
- W2895589541 hasPublicationYear "2019" @default.
- W2895589541 type Work @default.
- W2895589541 sameAs 2895589541 @default.
- W2895589541 citedByCount "7" @default.
- W2895589541 countsByYear W28955895412019 @default.
- W2895589541 countsByYear W28955895412020 @default.
- W2895589541 countsByYear W28955895412021 @default.
- W2895589541 countsByYear W28955895412022 @default.
- W2895589541 countsByYear W28955895412023 @default.
- W2895589541 crossrefType "journal-article" @default.
- W2895589541 hasAuthorship W2895589541A5003583859 @default.
- W2895589541 hasAuthorship W2895589541A5031755593 @default.
- W2895589541 hasAuthorship W2895589541A5049254107 @default.
- W2895589541 hasAuthorship W2895589541A5072203134 @default.
- W2895589541 hasAuthorship W2895589541A5072690470 @default.
- W2895589541 hasAuthorship W2895589541A5073070877 @default.
- W2895589541 hasAuthorship W2895589541A5082634513 @default.
- W2895589541 hasConcept C112972136 @default.
- W2895589541 hasConcept C115961682 @default.
- W2895589541 hasConcept C119857082 @default.
- W2895589541 hasConcept C153180895 @default.
- W2895589541 hasConcept C154945302 @default.
- W2895589541 hasConcept C184337299 @default.
- W2895589541 hasConcept C199360897 @default.
- W2895589541 hasConcept C2781238097 @default.
- W2895589541 hasConcept C41008148 @default.
- W2895589541 hasConcept C41608201 @default.
- W2895589541 hasConceptScore W2895589541C112972136 @default.
- W2895589541 hasConceptScore W2895589541C115961682 @default.
- W2895589541 hasConceptScore W2895589541C119857082 @default.
- W2895589541 hasConceptScore W2895589541C153180895 @default.
- W2895589541 hasConceptScore W2895589541C154945302 @default.
- W2895589541 hasConceptScore W2895589541C184337299 @default.
- W2895589541 hasConceptScore W2895589541C199360897 @default.
- W2895589541 hasConceptScore W2895589541C2781238097 @default.
- W2895589541 hasConceptScore W2895589541C41008148 @default.
- W2895589541 hasConceptScore W2895589541C41608201 @default.
- W2895589541 hasFunder F4320320698 @default.
- W2895589541 hasFunder F4320321001 @default.
- W2895589541 hasFunder F4320335787 @default.
- W2895589541 hasFunder F4320338464 @default.
- W2895589541 hasIssue "12" @default.
- W2895589541 hasLocation W28955895411 @default.
- W2895589541 hasLocation W28955895412 @default.
- W2895589541 hasOpenAccess W2895589541 @default.
- W2895589541 hasPrimaryLocation W28955895411 @default.
- W2895589541 hasRelatedWork W2092957489 @default.
- W2895589541 hasRelatedWork W2536452361 @default.
- W2895589541 hasRelatedWork W2893364174 @default.