Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313478851> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4313478851 endingPage "109292" @default.
- W4313478851 startingPage "109292" @default.
- W4313478851 abstract "Few shot segmentation (FSS) aims to learn pixel-level classification of a target object in a query image using only a few annotated support samples. This is challenging as it requires modeling appearance variations of target objects and the diverse visual cues between query and support images with limited information. To address this problem, we propose a semi-supervised FSS strategy that leverages additional prototypes from unlabeled images with uncertainty guided pseudo label refinement. To obtain reliable prototypes from unlabeled images, we meta-train a neural network to jointly predict segmentation and estimate the uncertainty of predictions. We employ the uncertainty estimates to exclude predictions with high degrees of uncertainty for pseudo label construction to obtain additional prototypes from the refined pseudo labels. During inference, query segmentation is predicted using prototypes from both support and unlabeled images including low-level features of the query images. Our approach can easily supplement existing approaches without the requirement of additional training when employing unlabeled samples. Extensive experiments on PASCAL-5i and COCO-20i demonstrate that our model can effectively remove unreliable predictions to refine pseudo labels and significantly improve upon baseline performance." @default.
- W4313478851 created "2023-01-06" @default.
- W4313478851 creator A5024112174 @default.
- W4313478851 creator A5025743587 @default.
- W4313478851 creator A5038323415 @default.
- W4313478851 creator A5076346519 @default.
- W4313478851 date "2023-05-01" @default.
- W4313478851 modified "2023-10-15" @default.
- W4313478851 title "Uncertainty-aware semi-supervised few shot segmentation" @default.
- W4313478851 cites W2117539524 @default.
- W4313478851 cites W2963078159 @default.
- W4313478851 cites W2964309882 @default.
- W4313478851 cites W3033502887 @default.
- W4313478851 cites W3034674637 @default.
- W4313478851 cites W3047258141 @default.
- W4313478851 cites W3108860939 @default.
- W4313478851 cites W3127022590 @default.
- W4313478851 cites W3173759468 @default.
- W4313478851 cites W3195778102 @default.
- W4313478851 cites W3213173905 @default.
- W4313478851 cites W4211244138 @default.
- W4313478851 cites W4280538209 @default.
- W4313478851 cites W4281768843 @default.
- W4313478851 doi "https://doi.org/10.1016/j.patcog.2022.109292" @default.
- W4313478851 hasPublicationYear "2023" @default.
- W4313478851 type Work @default.
- W4313478851 citedByCount "1" @default.
- W4313478851 countsByYear W43134788512023 @default.
- W4313478851 crossrefType "journal-article" @default.
- W4313478851 hasAuthorship W4313478851A5024112174 @default.
- W4313478851 hasAuthorship W4313478851A5025743587 @default.
- W4313478851 hasAuthorship W4313478851A5038323415 @default.
- W4313478851 hasAuthorship W4313478851A5076346519 @default.
- W4313478851 hasBestOaLocation W43134788512 @default.
- W4313478851 hasConcept C119857082 @default.
- W4313478851 hasConcept C124504099 @default.
- W4313478851 hasConcept C153180895 @default.
- W4313478851 hasConcept C154945302 @default.
- W4313478851 hasConcept C199360897 @default.
- W4313478851 hasConcept C2776214188 @default.
- W4313478851 hasConcept C41008148 @default.
- W4313478851 hasConcept C50644808 @default.
- W4313478851 hasConcept C75608658 @default.
- W4313478851 hasConcept C89600930 @default.
- W4313478851 hasConceptScore W4313478851C119857082 @default.
- W4313478851 hasConceptScore W4313478851C124504099 @default.
- W4313478851 hasConceptScore W4313478851C153180895 @default.
- W4313478851 hasConceptScore W4313478851C154945302 @default.
- W4313478851 hasConceptScore W4313478851C199360897 @default.
- W4313478851 hasConceptScore W4313478851C2776214188 @default.
- W4313478851 hasConceptScore W4313478851C41008148 @default.
- W4313478851 hasConceptScore W4313478851C50644808 @default.
- W4313478851 hasConceptScore W4313478851C75608658 @default.
- W4313478851 hasConceptScore W4313478851C89600930 @default.
- W4313478851 hasFunder F4320321681 @default.
- W4313478851 hasFunder F4320326308 @default.
- W4313478851 hasLocation W43134788511 @default.
- W4313478851 hasLocation W43134788512 @default.
- W4313478851 hasOpenAccess W4313478851 @default.
- W4313478851 hasPrimaryLocation W43134788511 @default.
- W4313478851 hasRelatedWork W134976887 @default.
- W4313478851 hasRelatedWork W1582206143 @default.
- W4313478851 hasRelatedWork W1840273037 @default.
- W4313478851 hasRelatedWork W2138214894 @default.
- W4313478851 hasRelatedWork W2549765251 @default.
- W4313478851 hasRelatedWork W2734888972 @default.
- W4313478851 hasRelatedWork W2972516102 @default.
- W4313478851 hasRelatedWork W3027394838 @default.
- W4313478851 hasRelatedWork W3161321444 @default.
- W4313478851 hasRelatedWork W4313854567 @default.
- W4313478851 hasVolume "137" @default.
- W4313478851 isParatext "false" @default.
- W4313478851 isRetracted "false" @default.
- W4313478851 workType "article" @default.