Matches in SemOpenAlex for { <https://semopenalex.org/work/W3043751495> ?p ?o ?g. }
- W3043751495 abstract "Existing techniques to encode spatial invariance within deep convolutional neural networks (CNNs) apply the same warping field to all the feature channels. This does not account for the fact that the individual feature channels can represent different semantic parts, which can undergo different spatial transformations w.r.t. a canonical configuration. To overcome this limitation, we introduce a learnable module, the volumetric transformer network (VTN), that predicts channel-wise warping fields so as to reconfigure intermediate CNN features spatially and channel-wisely. We design our VTN as an encoder-decoder network, with modules dedicated to letting the information flow across the feature channels, to account for the dependencies between the semantic parts. We further propose a loss function defined between the warped features of pairs of instances, which improves the localization ability of VTN. Our experiments show that VTN consistently boosts the features' representation power and consequently the networks' accuracy on fine-grained image recognition and instance-level image retrieval." @default.
- W3043751495 created "2020-07-23" @default.
- W3043751495 creator A5049300388 @default.
- W3043751495 creator A5078201467 @default.
- W3043751495 creator A5085363061 @default.
- W3043751495 date "2020-07-18" @default.
- W3043751495 modified "2023-10-14" @default.
- W3043751495 title "Volumetric Transformer Networks" @default.
- W3043751495 cites W1514535095 @default.
- W3043751495 cites W1833123814 @default.
- W3043751495 cites W1836465849 @default.
- W3043751495 cites W1846799578 @default.
- W3043751495 cites W1869500417 @default.
- W3043751495 cites W1936750108 @default.
- W3043751495 cites W1976794880 @default.
- W3043751495 cites W1993309459 @default.
- W3043751495 cites W2037227137 @default.
- W3043751495 cites W2103924867 @default.
- W3043751495 cites W2104657103 @default.
- W3043751495 cites W2128272608 @default.
- W3043751495 cites W2131846894 @default.
- W3043751495 cites W2138011018 @default.
- W3043751495 cites W2141362318 @default.
- W3043751495 cites W2141399712 @default.
- W3043751495 cites W2148809531 @default.
- W3043751495 cites W2151103935 @default.
- W3043751495 cites W2163605009 @default.
- W3043751495 cites W2177274842 @default.
- W3043751495 cites W2194775991 @default.
- W3043751495 cites W2202499615 @default.
- W3043751495 cites W2295107390 @default.
- W3043751495 cites W2295537791 @default.
- W3043751495 cites W2320444803 @default.
- W3043751495 cites W2544587078 @default.
- W3043751495 cites W2550553598 @default.
- W3043751495 cites W2552391307 @default.
- W3043751495 cites W2562066862 @default.
- W3043751495 cites W2601564443 @default.
- W3043751495 cites W2604195031 @default.
- W3043751495 cites W2613718673 @default.
- W3043751495 cites W2737725206 @default.
- W3043751495 cites W2740620254 @default.
- W3043751495 cites W2773003563 @default.
- W3043751495 cites W2797977484 @default.
- W3043751495 cites W2798365843 @default.
- W3043751495 cites W2804078698 @default.
- W3043751495 cites W2807931652 @default.
- W3043751495 cites W2883805248 @default.
- W3043751495 cites W2884585870 @default.
- W3043751495 cites W2889469641 @default.
- W3043751495 cites W2899771611 @default.
- W3043751495 cites W2916798096 @default.
- W3043751495 cites W2953334571 @default.
- W3043751495 cites W2955058313 @default.
- W3043751495 cites W2961018736 @default.
- W3043751495 cites W2962741254 @default.
- W3043751495 cites W2962761264 @default.
- W3043751495 cites W2962835968 @default.
- W3043751495 cites W2962858109 @default.
- W3043751495 cites W2963020784 @default.
- W3043751495 cites W2963066927 @default.
- W3043751495 cites W2963154697 @default.
- W3043751495 cites W2963203586 @default.
- W3043751495 cites W2963403868 @default.
- W3043751495 cites W2963407932 @default.
- W3043751495 cites W2963420686 @default.
- W3043751495 cites W2963446712 @default.
- W3043751495 cites W2963495494 @default.
- W3043751495 cites W2963564809 @default.
- W3043751495 cites W2963588253 @default.
- W3043751495 cites W2963727650 @default.
- W3043751495 cites W2963749936 @default.
- W3043751495 cites W2963966654 @default.
- W3043751495 cites W2964157791 @default.
- W3043751495 cites W2964308564 @default.
- W3043751495 cites W2982420602 @default.
- W3043751495 cites W2987761193 @default.
- W3043751495 cites W603908379 @default.
- W3043751495 cites W14333344 @default.
- W3043751495 doi "https://doi.org/10.48550/arxiv.2007.09433" @default.
- W3043751495 hasPublicationYear "2020" @default.
- W3043751495 type Work @default.
- W3043751495 sameAs 3043751495 @default.
- W3043751495 citedByCount "0" @default.
- W3043751495 crossrefType "posted-content" @default.
- W3043751495 hasAuthorship W3043751495A5049300388 @default.
- W3043751495 hasAuthorship W3043751495A5078201467 @default.
- W3043751495 hasAuthorship W3043751495A5085363061 @default.
- W3043751495 hasBestOaLocation W30437514951 @default.
- W3043751495 hasConcept C104317684 @default.
- W3043751495 hasConcept C108583219 @default.
- W3043751495 hasConcept C111919701 @default.
- W3043751495 hasConcept C118505674 @default.
- W3043751495 hasConcept C121332964 @default.
- W3043751495 hasConcept C127162648 @default.
- W3043751495 hasConcept C138885662 @default.
- W3043751495 hasConcept C153180895 @default.
- W3043751495 hasConcept C154945302 @default.
- W3043751495 hasConcept C157202957 @default.
- W3043751495 hasConcept C165801399 @default.