Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950790723> ?p ?o ?g. }
- W2950790723 abstract "Panoptic segmentation requires segments of both things (countable object instances) and stuff (uncountable and amorphous regions) within a single output. A common approach involves the fusion of instance segmentation (for things) and semantic segmentation (for stuff) into a non-overlapping placement of segments, and resolves overlaps. However, instance ordering with detection confidence do not correlate well with natural occlusion relationship. To resolve this issue, we propose a branch that is tasked with modeling how two instance masks should overlap one another as a binary relation. Our method, named OCFusion, is lightweight but particularly effective in the instance fusion process. OCFusion is trained with the ground truth relation derived automatically from the existing dataset annotations. We obtain state-of-the-art results on COCO and show competitive results on the Cityscapes panoptic segmentation benchmark." @default.
- W2950790723 created "2019-06-27" @default.
- W2950790723 creator A5001760915 @default.
- W2950790723 creator A5020532194 @default.
- W2950790723 creator A5052778038 @default.
- W2950790723 creator A5088583935 @default.
- W2950790723 date "2019-06-13" @default.
- W2950790723 modified "2023-09-23" @default.
- W2950790723 title "Learning Instance Occlusion for Panoptic Segmentation" @default.
- W2950790723 cites W1528789833 @default.
- W2950790723 cites W1530699444 @default.
- W2950790723 cites W1836465849 @default.
- W2950790723 cites W1861492603 @default.
- W2950790723 cites W1903029394 @default.
- W2950790723 cites W1927486677 @default.
- W2950790723 cites W19582845 @default.
- W2950790723 cites W1988790447 @default.
- W2950790723 cites W2031489346 @default.
- W2950790723 cites W2056860348 @default.
- W2950790723 cites W2076874408 @default.
- W2950790723 cites W2099972257 @default.
- W2950790723 cites W2103380420 @default.
- W2950790723 cites W2106624428 @default.
- W2950790723 cites W2107422845 @default.
- W2950790723 cites W2122006243 @default.
- W2950790723 cites W2124592697 @default.
- W2950790723 cites W2147800946 @default.
- W2950790723 cites W2156406284 @default.
- W2950790723 cites W2159386181 @default.
- W2950790723 cites W2162253476 @default.
- W2950790723 cites W2162846286 @default.
- W2950790723 cites W2216125271 @default.
- W2950790723 cites W2223259665 @default.
- W2950790723 cites W2322480645 @default.
- W2950790723 cites W2340897893 @default.
- W2950790723 cites W2412782625 @default.
- W2950790723 cites W2460807593 @default.
- W2950790723 cites W2536208356 @default.
- W2950790723 cites W2548197316 @default.
- W2950790723 cites W2557310260 @default.
- W2950790723 cites W2560023338 @default.
- W2950790723 cites W2565639579 @default.
- W2950790723 cites W2601564443 @default.
- W2950790723 cites W2612624696 @default.
- W2950790723 cites W2891601902 @default.
- W2950790723 cites W2902499724 @default.
- W2950790723 cites W2910628332 @default.
- W2950790723 cites W2911964244 @default.
- W2950790723 cites W2920326761 @default.
- W2950790723 cites W2950053619 @default.
- W2950790723 cites W2951638509 @default.
- W2950790723 cites W2953106684 @default.
- W2950790723 cites W2963150697 @default.
- W2950790723 cites W2963236837 @default.
- W2950790723 cites W2963393688 @default.
- W2950790723 cites W2963455537 @default.
- W2950790723 cites W2963660453 @default.
- W2950790723 cites W2963681621 @default.
- W2950790723 cites W2965182628 @default.
- W2950790723 cites W2974371958 @default.
- W2950790723 cites W2991405684 @default.
- W2950790723 cites W2999219213 @default.
- W2950790723 cites W809122546 @default.
- W2950790723 doi "https://doi.org/10.48550/arxiv.1906.05896" @default.
- W2950790723 hasPublicationYear "2019" @default.
- W2950790723 type Work @default.
- W2950790723 sameAs 2950790723 @default.
- W2950790723 citedByCount "6" @default.
- W2950790723 countsByYear W29507907232019 @default.
- W2950790723 countsByYear W29507907232020 @default.
- W2950790723 crossrefType "posted-content" @default.
- W2950790723 hasAuthorship W2950790723A5001760915 @default.
- W2950790723 hasAuthorship W2950790723A5020532194 @default.
- W2950790723 hasAuthorship W2950790723A5052778038 @default.
- W2950790723 hasAuthorship W2950790723A5088583935 @default.
- W2950790723 hasBestOaLocation W29507907231 @default.
- W2950790723 hasConcept C124101348 @default.
- W2950790723 hasConcept C13280743 @default.
- W2950790723 hasConcept C146849305 @default.
- W2950790723 hasConcept C153180895 @default.
- W2950790723 hasConcept C154945302 @default.
- W2950790723 hasConcept C185798385 @default.
- W2950790723 hasConcept C205649164 @default.
- W2950790723 hasConcept C25343380 @default.
- W2950790723 hasConcept C2781238097 @default.
- W2950790723 hasConcept C31972630 @default.
- W2950790723 hasConcept C41008148 @default.
- W2950790723 hasConcept C89600930 @default.
- W2950790723 hasConceptScore W2950790723C124101348 @default.
- W2950790723 hasConceptScore W2950790723C13280743 @default.
- W2950790723 hasConceptScore W2950790723C146849305 @default.
- W2950790723 hasConceptScore W2950790723C153180895 @default.
- W2950790723 hasConceptScore W2950790723C154945302 @default.
- W2950790723 hasConceptScore W2950790723C185798385 @default.
- W2950790723 hasConceptScore W2950790723C205649164 @default.
- W2950790723 hasConceptScore W2950790723C25343380 @default.
- W2950790723 hasConceptScore W2950790723C2781238097 @default.
- W2950790723 hasConceptScore W2950790723C31972630 @default.
- W2950790723 hasConceptScore W2950790723C41008148 @default.
- W2950790723 hasConceptScore W2950790723C89600930 @default.