FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3111551570> ?p ?o ?g. }

• W3111551570 abstract "Designed to learn long-range interactions on sequential data, transformers continue to show state-of-the-art results on a wide variety of tasks. In contrast to CNNs, they contain no inductive bias that prioritizes local interactions. This makes them expressive, but also computationally infeasible for long sequences, such as high-resolution images. We demonstrate how combining the effectiveness of the inductive bias of CNNs with the expressivity of transformers enables them to model and thereby synthesize high-resolution images. We show how to (i) use CNNs to learn a context-rich vocabulary of image constituents, and in turn (ii) utilize transformers to efficiently model their composition within high-resolution images. Our approach is readily applied to conditional synthesis tasks, where both non-spatial information, such as object classes, and spatial information, such as segmentations, can control the generated image. In particular, we present the first results on semantically-guided synthesis of megapixel images with transformers and obtain the state of the art among autoregressive models on class-conditional ImageNet. Code and pretrained models can be found at this https URL ." @default.
• W3111551570 created "2020-12-21" @default.
• W3111551570 creator A5022021439 @default.
• W3111551570 creator A5051842958 @default.
• W3111551570 creator A5084415727 @default.
• W3111551570 date "2020-12-17" @default.
• W3111551570 modified "2023-09-27" @default.
• W3111551570 title "Taming Transformers for High-Resolution Image Synthesis" @default.
• W3111551570 cites W1959608418 @default.
• W3111551570 cites W2099471712 @default.
• W3111551570 cites W2108598243 @default.
• W3111551570 cites W2171108400 @default.
• W3111551570 cites W2194775991 @default.
• W3111551570 cites W2242818861 @default.
• W3111551570 cites W2267126114 @default.
• W3111551570 cites W2269752429 @default.
• W3111551570 cites W2331128040 @default.
• W3111551570 cites W2348664362 @default.
• W3111551570 cites W2412782625 @default.
• W3111551570 cites W2413794162 @default.
• W3111551570 cites W2471768434 @default.
• W3111551570 cites W2507296351 @default.
• W3111551570 cites W2547875792 @default.
• W3111551570 cites W2881214865 @default.
• W3111551570 cites W2893749619 @default.
• W3111551570 cites W2902630600 @default.
• W3111551570 cites W2922386270 @default.
• W3111551570 cites W2940744433 @default.
• W3111551570 cites W2945924057 @default.
• W3111551570 cites W2949847915 @default.
• W3111551570 cites W2962760235 @default.
• W3111551570 cites W2962770929 @default.
• W3111551570 cites W2962785568 @default.
• W3111551570 cites W2962820504 @default.
• W3111551570 cites W2962897886 @default.
• W3111551570 cites W2962974533 @default.
• W3111551570 cites W2963045354 @default.
• W3111551570 cites W2963073614 @default.
• W3111551570 cites W2963139417 @default.
• W3111551570 cites W2963174698 @default.
• W3111551570 cites W2963403868 @default.
• W3111551570 cites W2963428348 @default.
• W3111551570 cites W2963522749 @default.
• W3111551570 cites W2963799213 @default.
• W3111551570 cites W2963800363 @default.
• W3111551570 cites W2964308564 @default.
• W3111551570 cites W2970315999 @default.
• W3111551570 cites W2970718407 @default.
• W3111551570 cites W2971074500 @default.
• W3111551570 cites W2971480596 @default.
• W3111551570 cites W2981721547 @default.
• W3111551570 cites W2994760783 @default.
• W3111551570 cites W2996287690 @default.
• W3111551570 cites W2998108143 @default.
• W3111551570 cites W3025973238 @default.
• W3111551570 cites W3030163527 @default.
• W3111551570 cites W3034625979 @default.
• W3111551570 cites W3034723751 @default.
• W3111551570 cites W3034776267 @default.
• W3111551570 cites W3035574324 @default.
• W3111551570 cites W3035687950 @default.
• W3111551570 cites W3081167590 @default.
• W3111551570 cites W3083914148 @default.
• W3111551570 cites W3092033429 @default.
• W3111551570 cites W3098510582 @default.
• W3111551570 cites W3099188186 @default.
• W3111551570 cites W3103781353 @default.
• W3111551570 cites W3104876213 @default.
• W3111551570 cites W3106570356 @default.
• W3111551570 cites W3108329879 @default.
• W3111551570 cites W3118605064 @default.
• W3111551570 cites W3162926177 @default.
• W3111551570 cites W3176823897 @default.
• W3111551570 cites W967544008 @default.
• W3111551570 hasPublicationYear "2020" @default.
• W3111551570 type Work @default.
• W3111551570 sameAs 3111551570 @default.
• W3111551570 citedByCount "18" @default.
• W3111551570 countsByYear W31115515702021 @default.
• W3111551570 crossrefType "posted-content" @default.
• W3111551570 hasAuthorship W3111551570A5022021439 @default.
• W3111551570 hasAuthorship W3111551570A5051842958 @default.
• W3111551570 hasAuthorship W3111551570A5084415727 @default.
• W3111551570 hasConcept C121332964 @default.
• W3111551570 hasConcept C138885662 @default.
• W3111551570 hasConcept C149782125 @default.
• W3111551570 hasConcept C153180895 @default.
• W3111551570 hasConcept C154945302 @default.
• W3111551570 hasConcept C159877910 @default.
• W3111551570 hasConcept C162324750 @default.
• W3111551570 hasConcept C165801399 @default.
• W3111551570 hasConcept C187736073 @default.
• W3111551570 hasConcept C197352929 @default.
• W3111551570 hasConcept C2777601683 @default.
• W3111551570 hasConcept C2780451532 @default.
• W3111551570 hasConcept C28006648 @default.
• W3111551570 hasConcept C33923547 @default.
• W3111551570 hasConcept C41008148 @default.
• W3111551570 hasConcept C41895202 @default.
• W3111551570 hasConcept C62520636 @default.

• next