Matches in SemOpenAlex for { <https://semopenalex.org/work/W3203383669> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3203383669 endingPage "32" @default.
- W3203383669 startingPage "18" @default.
- W3203383669 abstract "We propose a novel method for unsupervised semantic image segmentation based on mutual information maximization between local and global high-level image features. The core idea of our work is to leverage recent progress in self-supervised image representation learning. Representation learning methods compute a single high-level feature capturing an entire image. In contrast, we compute multiple high-level features, each capturing image segments of one particular semantic class. To this end, we propose a novel two-step learning procedure comprising a segmentation and a mutual information maximization step. In the first step, we segment images based on local and global features. In the second step, we maximize the mutual information between local features and high-level features of their respective class. For training, we provide solely unlabeled images and start from random network initialization. For quantitative and qualitative evaluation, we use established benchmarks, and COCO-Persons, whereby we introduce the latter in this paper as a challenging novel benchmark. InfoSeg significantly outperforms the current state-of-the-art, e.g., we achieve a relative increase of (26%) in the Pixel Accuracy metric on the COCO-Stuff dataset." @default.
- W3203383669 created "2021-10-11" @default.
- W3203383669 creator A5029214820 @default.
- W3203383669 creator A5069159614 @default.
- W3203383669 date "2021-01-01" @default.
- W3203383669 modified "2023-10-14" @default.
- W3203383669 title "InfoSeg: Unsupervised Semantic Image Segmentation with Mutual Information Maximization" @default.
- W3203383669 cites W1861492603 @default.
- W3203383669 cites W1999478155 @default.
- W3203383669 cites W2067191022 @default.
- W3203383669 cites W2108598243 @default.
- W3203383669 cites W2110158442 @default.
- W3203383669 cites W2121947440 @default.
- W3203383669 cites W2222512263 @default.
- W3203383669 cites W2561196672 @default.
- W3203383669 cites W2607333215 @default.
- W3203383669 cites W2798991696 @default.
- W3203383669 cites W2982106100 @default.
- W3203383669 cites W2990500698 @default.
- W3203383669 cites W3035524453 @default.
- W3203383669 cites W3107725261 @default.
- W3203383669 cites W3110446398 @default.
- W3203383669 cites W3181191249 @default.
- W3203383669 cites W343636949 @default.
- W3203383669 doi "https://doi.org/10.1007/978-3-030-92659-5_2" @default.
- W3203383669 hasPublicationYear "2021" @default.
- W3203383669 type Work @default.
- W3203383669 sameAs 3203383669 @default.
- W3203383669 citedByCount "7" @default.
- W3203383669 countsByYear W32033836692022 @default.
- W3203383669 countsByYear W32033836692023 @default.
- W3203383669 crossrefType "book-chapter" @default.
- W3203383669 hasAuthorship W3203383669A5029214820 @default.
- W3203383669 hasAuthorship W3203383669A5069159614 @default.
- W3203383669 hasBestOaLocation W32033836692 @default.
- W3203383669 hasConcept C114466953 @default.
- W3203383669 hasConcept C115961682 @default.
- W3203383669 hasConcept C119857082 @default.
- W3203383669 hasConcept C124504099 @default.
- W3203383669 hasConcept C126255220 @default.
- W3203383669 hasConcept C138885662 @default.
- W3203383669 hasConcept C152139883 @default.
- W3203383669 hasConcept C153083717 @default.
- W3203383669 hasConcept C153180895 @default.
- W3203383669 hasConcept C154945302 @default.
- W3203383669 hasConcept C199360897 @default.
- W3203383669 hasConcept C2776330181 @default.
- W3203383669 hasConcept C2776401178 @default.
- W3203383669 hasConcept C33923547 @default.
- W3203383669 hasConcept C41008148 @default.
- W3203383669 hasConcept C41895202 @default.
- W3203383669 hasConcept C59404180 @default.
- W3203383669 hasConcept C89600930 @default.
- W3203383669 hasConceptScore W3203383669C114466953 @default.
- W3203383669 hasConceptScore W3203383669C115961682 @default.
- W3203383669 hasConceptScore W3203383669C119857082 @default.
- W3203383669 hasConceptScore W3203383669C124504099 @default.
- W3203383669 hasConceptScore W3203383669C126255220 @default.
- W3203383669 hasConceptScore W3203383669C138885662 @default.
- W3203383669 hasConceptScore W3203383669C152139883 @default.
- W3203383669 hasConceptScore W3203383669C153083717 @default.
- W3203383669 hasConceptScore W3203383669C153180895 @default.
- W3203383669 hasConceptScore W3203383669C154945302 @default.
- W3203383669 hasConceptScore W3203383669C199360897 @default.
- W3203383669 hasConceptScore W3203383669C2776330181 @default.
- W3203383669 hasConceptScore W3203383669C2776401178 @default.
- W3203383669 hasConceptScore W3203383669C33923547 @default.
- W3203383669 hasConceptScore W3203383669C41008148 @default.
- W3203383669 hasConceptScore W3203383669C41895202 @default.
- W3203383669 hasConceptScore W3203383669C59404180 @default.
- W3203383669 hasConceptScore W3203383669C89600930 @default.
- W3203383669 hasLocation W32033836691 @default.
- W3203383669 hasLocation W32033836692 @default.
- W3203383669 hasOpenAccess W3203383669 @default.
- W3203383669 hasPrimaryLocation W32033836691 @default.
- W3203383669 hasRelatedWork W1566144976 @default.
- W3203383669 hasRelatedWork W1983106178 @default.
- W3203383669 hasRelatedWork W2114054925 @default.
- W3203383669 hasRelatedWork W2241417046 @default.
- W3203383669 hasRelatedWork W2393900892 @default.
- W3203383669 hasRelatedWork W2597403208 @default.
- W3203383669 hasRelatedWork W2908598672 @default.
- W3203383669 hasRelatedWork W3176564347 @default.
- W3203383669 hasRelatedWork W4306760398 @default.
- W3203383669 hasRelatedWork W4307206503 @default.
- W3203383669 isParatext "false" @default.
- W3203383669 isRetracted "false" @default.
- W3203383669 magId "3203383669" @default.
- W3203383669 workType "book-chapter" @default.