Matches in SemOpenAlex for { <https://semopenalex.org/work/W3134428853> ?p ?o ?g. }
- W3134428853 abstract "Superpixels are a useful representation to reduce the complexity of image data. However, to combine superpixels with convolutional neural networks (CNNs) in an end-to-end fashion, one requires extra models to generate superpixels and special operations such as graph convolution. In this paper, we propose a way to implicitly integrate a superpixel scheme into CNNs, which makes it easy to use superpixels with CNNs in an end-to-end fashion. Our proposed method hierarchically groups pixels at downsampling layers and generates superpixels. Our method can be plugged into many existing architectures without a change in their feed-forward path because our method does not use superpixels in the feed-forward path but use them to recover the lost resolution instead of bilinear upsampling. As a result, our method preserves detailed information such as object boundaries in the form of superpixels even when the model contains downsampling layers. We evaluate our method on several tasks such as semantic segmentation, superpixel segmentation, and monocular depth estimation, and confirm that it speeds up modern architectures and/or improves their prediction accuracy in these tasks." @default.
- W3134428853 created "2021-03-15" @default.
- W3134428853 creator A5023177033 @default.
- W3134428853 date "2021-03-04" @default.
- W3134428853 modified "2023-09-25" @default.
- W3134428853 title "Implicit Integration of Superpixel Segmentation into Fully Convolutional Networks" @default.
- W3134428853 cites W125693051 @default.
- W3134428853 cites W1745334888 @default.
- W3134428853 cites W1901129140 @default.
- W3134428853 cites W1903029394 @default.
- W3134428853 cites W2118246710 @default.
- W3134428853 cites W2121927366 @default.
- W3134428853 cites W2172859194 @default.
- W3134428853 cites W2194775991 @default.
- W3134428853 cites W2340897893 @default.
- W3134428853 cites W2412782625 @default.
- W3134428853 cites W2419448466 @default.
- W3134428853 cites W2507296351 @default.
- W3134428853 cites W2547875792 @default.
- W3134428853 cites W2557406251 @default.
- W3134428853 cites W2558460151 @default.
- W3134428853 cites W2560023338 @default.
- W3134428853 cites W2560474170 @default.
- W3134428853 cites W2560533888 @default.
- W3134428853 cites W2565639579 @default.
- W3134428853 cites W2585293115 @default.
- W3134428853 cites W2605214291 @default.
- W3134428853 cites W2630837129 @default.
- W3134428853 cites W2738588019 @default.
- W3134428853 cites W2798380924 @default.
- W3134428853 cites W2883606943 @default.
- W3134428853 cites W2892064179 @default.
- W3134428853 cites W2910628332 @default.
- W3134428853 cites W2945767858 @default.
- W3134428853 cites W2952165242 @default.
- W3134428853 cites W2962837172 @default.
- W3134428853 cites W2963091558 @default.
- W3134428853 cites W2963372527 @default.
- W3134428853 cites W2963488291 @default.
- W3134428853 cites W2963782415 @default.
- W3134428853 cites W2963840672 @default.
- W3134428853 cites W2963881378 @default.
- W3134428853 cites W2963917315 @default.
- W3134428853 cites W2964030969 @default.
- W3134428853 cites W2964121744 @default.
- W3134428853 cites W2964288706 @default.
- W3134428853 cites W2964309882 @default.
- W3134428853 cites W2970971581 @default.
- W3134428853 cites W2985775862 @default.
- W3134428853 cites W2991471181 @default.
- W3134428853 cites W3015575917 @default.
- W3134428853 cites W3022478135 @default.
- W3134428853 cites W3035241330 @default.
- W3134428853 cites W3110440461 @default.
- W3134428853 cites W764651262 @default.
- W3134428853 doi "https://doi.org/10.48550/arxiv.2103.03435" @default.
- W3134428853 hasPublicationYear "2021" @default.
- W3134428853 type Work @default.
- W3134428853 sameAs 3134428853 @default.
- W3134428853 citedByCount "0" @default.
- W3134428853 crossrefType "posted-content" @default.
- W3134428853 hasAuthorship W3134428853A5023177033 @default.
- W3134428853 hasBestOaLocation W31344288531 @default.
- W3134428853 hasConcept C110384440 @default.
- W3134428853 hasConcept C115961682 @default.
- W3134428853 hasConcept C132525143 @default.
- W3134428853 hasConcept C153180895 @default.
- W3134428853 hasConcept C154945302 @default.
- W3134428853 hasConcept C160633673 @default.
- W3134428853 hasConcept C199360897 @default.
- W3134428853 hasConcept C205203396 @default.
- W3134428853 hasConcept C2777735758 @default.
- W3134428853 hasConcept C31972630 @default.
- W3134428853 hasConcept C41008148 @default.
- W3134428853 hasConcept C45347329 @default.
- W3134428853 hasConcept C50644808 @default.
- W3134428853 hasConcept C80444323 @default.
- W3134428853 hasConcept C81363708 @default.
- W3134428853 hasConcept C89600930 @default.
- W3134428853 hasConceptScore W3134428853C110384440 @default.
- W3134428853 hasConceptScore W3134428853C115961682 @default.
- W3134428853 hasConceptScore W3134428853C132525143 @default.
- W3134428853 hasConceptScore W3134428853C153180895 @default.
- W3134428853 hasConceptScore W3134428853C154945302 @default.
- W3134428853 hasConceptScore W3134428853C160633673 @default.
- W3134428853 hasConceptScore W3134428853C199360897 @default.
- W3134428853 hasConceptScore W3134428853C205203396 @default.
- W3134428853 hasConceptScore W3134428853C2777735758 @default.
- W3134428853 hasConceptScore W3134428853C31972630 @default.
- W3134428853 hasConceptScore W3134428853C41008148 @default.
- W3134428853 hasConceptScore W3134428853C45347329 @default.
- W3134428853 hasConceptScore W3134428853C50644808 @default.
- W3134428853 hasConceptScore W3134428853C80444323 @default.
- W3134428853 hasConceptScore W3134428853C81363708 @default.
- W3134428853 hasConceptScore W3134428853C89600930 @default.
- W3134428853 hasLocation W31344288531 @default.
- W3134428853 hasOpenAccess W3134428853 @default.
- W3134428853 hasPrimaryLocation W31344288531 @default.
- W3134428853 hasRelatedWork W1669643531 @default.
- W3134428853 hasRelatedWork W2005437358 @default.