Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385805000> ?p ?o ?g. }
- W4385805000 abstract "Most dense recognition approaches bring a separate decision in each particular pixel. These approaches deliver competitive performance in usual closed-set setups. However, important applications in the wild typically require strong performance in presence of outliers. We show that this demanding setup greatly benefits from mask-level predictions, even in the case of non-finetuned baseline models. Moreover, we propose an alternative formulation of dense recognition uncertainty that effectively reduces false positive responses at semantic borders. The proposed formulation produces a further improvement over a very strong baseline and sets the new state of the art in outlier-aware semantic segmentation with and without training on negative data. Our contributions also lead to performance improvement in a recent panoptic setup. In-depth experiments confirm that our approach succeeds due to implicit aggregation of pixel-level cues into mask-level predictions." @default.
- W4385805000 created "2023-08-15" @default.
- W4385805000 creator A5032164394 @default.
- W4385805000 creator A5073708578 @default.
- W4385805000 creator A5083625987 @default.
- W4385805000 date "2023-06-01" @default.
- W4385805000 modified "2023-09-26" @default.
- W4385805000 title "On Advantages of Mask-level Recognition for Outlier-aware Segmentation" @default.
- W4385805000 cites W1507506748 @default.
- W4385805000 cites W1903029394 @default.
- W4385805000 cites W1923115158 @default.
- W4385805000 cites W2018459374 @default.
- W4385805000 cites W2022508996 @default.
- W4385805000 cites W2119880843 @default.
- W4385805000 cites W2322480645 @default.
- W4385805000 cites W2340897893 @default.
- W4385805000 cites W2507296351 @default.
- W4385805000 cites W2781228439 @default.
- W4385805000 cites W2894903065 @default.
- W4385805000 cites W2933610837 @default.
- W4385805000 cites W2963073398 @default.
- W4385805000 cites W2963149653 @default.
- W4385805000 cites W2981786146 @default.
- W4385805000 cites W2982362232 @default.
- W4385805000 cites W3011680720 @default.
- W4385805000 cites W3034355852 @default.
- W4385805000 cites W3035358681 @default.
- W4385805000 cites W3077024029 @default.
- W4385805000 cites W3090238777 @default.
- W4385805000 cites W3110437118 @default.
- W4385805000 cites W3112906266 @default.
- W4385805000 cites W3129166376 @default.
- W4385805000 cites W3130453925 @default.
- W4385805000 cites W3138516171 @default.
- W4385805000 cites W3176640650 @default.
- W4385805000 cites W3177266689 @default.
- W4385805000 cites W3193198176 @default.
- W4385805000 cites W3201876338 @default.
- W4385805000 cites W3204557602 @default.
- W4385805000 cites W4214530306 @default.
- W4385805000 cites W4281666638 @default.
- W4385805000 cites W4312601785 @default.
- W4385805000 cites W4312815172 @default.
- W4385805000 cites W78159342 @default.
- W4385805000 doi "https://doi.org/10.1109/cvprw59228.2023.00295" @default.
- W4385805000 hasPublicationYear "2023" @default.
- W4385805000 type Work @default.
- W4385805000 citedByCount "0" @default.
- W4385805000 crossrefType "proceedings-article" @default.
- W4385805000 hasAuthorship W4385805000A5032164394 @default.
- W4385805000 hasAuthorship W4385805000A5073708578 @default.
- W4385805000 hasAuthorship W4385805000A5083625987 @default.
- W4385805000 hasBestOaLocation W43858050002 @default.
- W4385805000 hasConcept C111368507 @default.
- W4385805000 hasConcept C119857082 @default.
- W4385805000 hasConcept C12725497 @default.
- W4385805000 hasConcept C127313418 @default.
- W4385805000 hasConcept C153180895 @default.
- W4385805000 hasConcept C154945302 @default.
- W4385805000 hasConcept C160633673 @default.
- W4385805000 hasConcept C162324750 @default.
- W4385805000 hasConcept C177264268 @default.
- W4385805000 hasConcept C199360897 @default.
- W4385805000 hasConcept C21547014 @default.
- W4385805000 hasConcept C2778915421 @default.
- W4385805000 hasConcept C41008148 @default.
- W4385805000 hasConcept C51632099 @default.
- W4385805000 hasConcept C739882 @default.
- W4385805000 hasConcept C79337645 @default.
- W4385805000 hasConcept C89600930 @default.
- W4385805000 hasConceptScore W4385805000C111368507 @default.
- W4385805000 hasConceptScore W4385805000C119857082 @default.
- W4385805000 hasConceptScore W4385805000C12725497 @default.
- W4385805000 hasConceptScore W4385805000C127313418 @default.
- W4385805000 hasConceptScore W4385805000C153180895 @default.
- W4385805000 hasConceptScore W4385805000C154945302 @default.
- W4385805000 hasConceptScore W4385805000C160633673 @default.
- W4385805000 hasConceptScore W4385805000C162324750 @default.
- W4385805000 hasConceptScore W4385805000C177264268 @default.
- W4385805000 hasConceptScore W4385805000C199360897 @default.
- W4385805000 hasConceptScore W4385805000C21547014 @default.
- W4385805000 hasConceptScore W4385805000C2778915421 @default.
- W4385805000 hasConceptScore W4385805000C41008148 @default.
- W4385805000 hasConceptScore W4385805000C51632099 @default.
- W4385805000 hasConceptScore W4385805000C739882 @default.
- W4385805000 hasConceptScore W4385805000C79337645 @default.
- W4385805000 hasConceptScore W4385805000C89600930 @default.
- W4385805000 hasLocation W43858050001 @default.
- W4385805000 hasLocation W43858050002 @default.
- W4385805000 hasOpenAccess W4385805000 @default.
- W4385805000 hasPrimaryLocation W43858050001 @default.
- W4385805000 hasRelatedWork W2051290323 @default.
- W4385805000 hasRelatedWork W2076520961 @default.
- W4385805000 hasRelatedWork W2085936892 @default.
- W4385805000 hasRelatedWork W2136485282 @default.
- W4385805000 hasRelatedWork W2402385015 @default.
- W4385805000 hasRelatedWork W2546871836 @default.
- W4385805000 hasRelatedWork W3137655554 @default.
- W4385805000 hasRelatedWork W3216541841 @default.
- W4385805000 hasRelatedWork W4213268118 @default.