Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312907595> ?p ?o ?g. }
- W4312907595 endingPage "505" @default.
- W4312907595 startingPage "487" @default.
- W4312907595 abstract "Integrating high-level context information with low-level details is of central importance in semantic segmentation. Towards this end, most existing segmentation models apply bilinear up-sampling and convolutions to feature maps of different scales, and then align them at the same resolution. However, bilinear up-sampling blurs the precise information learned in these feature maps and convolutions incur extra computation costs. To address these issues, we propose the Implicit Feature Alignment function (IFA). Our method is inspired by the rapidly expanding topic of implicit neural representations, where coordinate-based neural networks are used to designate fields of signals. In IFA, feature vectors are viewed as representing a 2D field of information. Given a query coordinate, nearby feature vectors with their relative coordinates are taken from the multi-level feature maps and then fed into an MLP to generate the corresponding output. As such, IFA implicitly aligns the feature maps at different levels and is capable of producing segmentation maps in arbitrary resolutions. We demonstrate the efficacy of IFA on multiple datasets, including Cityscapes, PASCAL Context, and ADE20K. Our method can be combined with improvement on various architectures, and it achieves state-of-the-art computation-accuracy trade-off on common benchmarks. Code is available at https://github.com/hzhupku/IFA ." @default.
- W4312907595 created "2023-01-05" @default.
- W4312907595 creator A5038550641 @default.
- W4312907595 creator A5048249044 @default.
- W4312907595 creator A5053652981 @default.
- W4312907595 creator A5063957844 @default.
- W4312907595 creator A5066028215 @default.
- W4312907595 creator A5087970989 @default.
- W4312907595 creator A5090180284 @default.
- W4312907595 date "2022-01-01" @default.
- W4312907595 modified "2023-10-16" @default.
- W4312907595 title "Learning Implicit Feature Alignment Function for Semantic Segmentation" @default.
- W4312907595 cites W1901129140 @default.
- W4312907595 cites W1903029394 @default.
- W4312907595 cites W2125215748 @default.
- W4312907595 cites W2194775991 @default.
- W4312907595 cites W2340897893 @default.
- W4312907595 cites W2412782625 @default.
- W4312907595 cites W2560023338 @default.
- W4312907595 cites W2563705555 @default.
- W4312907595 cites W2565639579 @default.
- W4312907595 cites W2737258237 @default.
- W4312907595 cites W2799166040 @default.
- W4312907595 cites W2799213142 @default.
- W4312907595 cites W2886934227 @default.
- W4312907595 cites W2888340395 @default.
- W4312907595 cites W2895340641 @default.
- W4312907595 cites W2910628332 @default.
- W4312907595 cites W2955058313 @default.
- W4312907595 cites W2955813853 @default.
- W4312907595 cites W2962849139 @default.
- W4312907595 cites W2963091558 @default.
- W4312907595 cites W2963516811 @default.
- W4312907595 cites W2963627347 @default.
- W4312907595 cites W2963727650 @default.
- W4312907595 cites W2963926543 @default.
- W4312907595 cites W2964309882 @default.
- W4312907595 cites W2965391153 @default.
- W4312907595 cites W2981657250 @default.
- W4312907595 cites W2981689412 @default.
- W4312907595 cites W2981899103 @default.
- W4312907595 cites W2981978060 @default.
- W4312907595 cites W2987322772 @default.
- W4312907595 cites W2989630530 @default.
- W4312907595 cites W2990775046 @default.
- W4312907595 cites W2993235622 @default.
- W4312907595 cites W3014641072 @default.
- W4312907595 cites W3035339581 @default.
- W4312907595 cites W3035515538 @default.
- W4312907595 cites W3107113572 @default.
- W4312907595 cites W3109196706 @default.
- W4312907595 cites W3109301572 @default.
- W4312907595 cites W3109585842 @default.
- W4312907595 cites W3110440461 @default.
- W4312907595 cites W3169770376 @default.
- W4312907595 cites W3170841864 @default.
- W4312907595 cites W3174865552 @default.
- W4312907595 cites W3202623927 @default.
- W4312907595 cites W4214893857 @default.
- W4312907595 cites W4312322023 @default.
- W4312907595 cites W4312472971 @default.
- W4312907595 doi "https://doi.org/10.1007/978-3-031-19818-2_28" @default.
- W4312907595 hasPublicationYear "2022" @default.
- W4312907595 type Work @default.
- W4312907595 citedByCount "9" @default.
- W4312907595 countsByYear W43129075952023 @default.
- W4312907595 crossrefType "book-chapter" @default.
- W4312907595 hasAuthorship W4312907595A5038550641 @default.
- W4312907595 hasAuthorship W4312907595A5048249044 @default.
- W4312907595 hasAuthorship W4312907595A5053652981 @default.
- W4312907595 hasAuthorship W4312907595A5063957844 @default.
- W4312907595 hasAuthorship W4312907595A5066028215 @default.
- W4312907595 hasAuthorship W4312907595A5087970989 @default.
- W4312907595 hasAuthorship W4312907595A5090180284 @default.
- W4312907595 hasBestOaLocation W43129075952 @default.
- W4312907595 hasConcept C11413529 @default.
- W4312907595 hasConcept C138885662 @default.
- W4312907595 hasConcept C151730666 @default.
- W4312907595 hasConcept C153180895 @default.
- W4312907595 hasConcept C154945302 @default.
- W4312907595 hasConcept C199360897 @default.
- W4312907595 hasConcept C205203396 @default.
- W4312907595 hasConcept C2776401178 @default.
- W4312907595 hasConcept C2779343474 @default.
- W4312907595 hasConcept C31972630 @default.
- W4312907595 hasConcept C41008148 @default.
- W4312907595 hasConcept C41895202 @default.
- W4312907595 hasConcept C45374587 @default.
- W4312907595 hasConcept C75608658 @default.
- W4312907595 hasConcept C83665646 @default.
- W4312907595 hasConcept C86803240 @default.
- W4312907595 hasConcept C89600930 @default.
- W4312907595 hasConceptScore W4312907595C11413529 @default.
- W4312907595 hasConceptScore W4312907595C138885662 @default.
- W4312907595 hasConceptScore W4312907595C151730666 @default.
- W4312907595 hasConceptScore W4312907595C153180895 @default.
- W4312907595 hasConceptScore W4312907595C154945302 @default.
- W4312907595 hasConceptScore W4312907595C199360897 @default.