FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3172425208> ?p ?o ?g. }

• W3172425208 abstract "The impressive performance of deep convolutional neural networks in single-view 3D reconstruction suggests that these models perform non-trivial reasoning about the 3D structure of the output space. Recent work has challenged this belief, showing that, on standard benchmarks, complex encoder-decoder architectures perform similarly to nearest-neighbor baselines or simple linear decoder models that exploit large amounts of per-category data. However, building large collections of 3D shapes for supervised training is a laborious process; a more realistic and less constraining task is inferring 3D shapes for categories with few available training examples, calling for a model that can successfully generalize to novel object classes. In this work we experimentally demonstrate that naive baselines fail in this few-shot learning setting, in which the network must learn informative shape priors for inference of new categories. We propose three ways to learn a class-specific global shape prior, directly from data. Using these techniques, we are able to capture multi-scale information about the 3D shape, and account for intra-class variability by virtue of an implicit compositional structure. Experiments on the popular ShapeNet dataset show that our method outperforms a zero-shot baseline by over 40%, and the current state-of-the-art by over 10%, in terms of relative performance, in the few-shot setting." @default.
• W3172425208 created "2021-06-22" @default.
• W3172425208 creator A5014648528 @default.
• W3172425208 creator A5020523405 @default.
• W3172425208 creator A5025113992 @default.
• W3172425208 creator A5040003361 @default.
• W3172425208 creator A5069883194 @default.
• W3172425208 creator A5087256918 @default.
• W3172425208 date "2021-06-11" @default.
• W3172425208 modified "2023-10-04" @default.
• W3172425208 title "Learning Compositional Shape Priors for Few-Shot 3D Reconstruction." @default.
• W3172425208 cites W1576579612 @default.
• W3172425208 cites W2013599012 @default.
• W3172425208 cites W2022156968 @default.
• W3172425208 cites W2027174336 @default.
• W3172425208 cites W2038763636 @default.
• W3172425208 cites W2039262381 @default.
• W3172425208 cites W2053825465 @default.
• W3172425208 cites W2074954154 @default.
• W3172425208 cites W2089468765 @default.
• W3172425208 cites W2121752747 @default.
• W3172425208 cites W2122676594 @default.
• W3172425208 cites W2151996626 @default.
• W3172425208 cites W2155196764 @default.
• W3172425208 cites W2168545424 @default.
• W3172425208 cites W2190691619 @default.
• W3172425208 cites W2194775991 @default.
• W3172425208 cites W2342277278 @default.
• W3172425208 cites W2546066744 @default.
• W3172425208 cites W2551540143 @default.
• W3172425208 cites W2601450892 @default.
• W3172425208 cites W2603429625 @default.
• W3172425208 cites W2604763608 @default.
• W3172425208 cites W2748512037 @default.
• W3172425208 cites W2768376748 @default.
• W3172425208 cites W2796346823 @default.
• W3172425208 cites W2893918048 @default.
• W3172425208 cites W2904383797 @default.
• W3172425208 cites W2914390273 @default.
• W3172425208 cites W2944579304 @default.
• W3172425208 cites W2947252423 @default.
• W3172425208 cites W2962778872 @default.
• W3172425208 cites W2962790997 @default.
• W3172425208 cites W2962849139 @default.
• W3172425208 cites W2963078860 @default.
• W3172425208 cites W2963123301 @default.
• W3172425208 cites W2963141648 @default.
• W3172425208 cites W2963341924 @default.
• W3172425208 cites W2963627347 @default.
• W3172425208 cites W2963641844 @default.
• W3172425208 cites W2963730200 @default.
• W3172425208 cites W2963845150 @default.
• W3172425208 cites W2963921132 @default.
• W3172425208 cites W2964121744 @default.
• W3172425208 cites W2964137676 @default.
• W3172425208 cites W2970899367 @default.
• W3172425208 cites W2982593143 @default.
• W3172425208 cites W2983582925 @default.
• W3172425208 cites W2994633389 @default.
• W3172425208 cites W2994886384 @default.
• W3172425208 cites W2997616671 @default.
• W3172425208 cites W3034858314 @default.
• W3172425208 cites W3035513921 @default.
• W3172425208 cites W3091905774 @default.
• W3172425208 cites W3096831136 @default.
• W3172425208 cites W3100456304 @default.
• W3172425208 cites W3106165820 @default.
• W3172425208 hasPublicationYear "2021" @default.
• W3172425208 type Work @default.
• W3172425208 sameAs 3172425208 @default.
• W3172425208 citedByCount "0" @default.
• W3172425208 crossrefType "posted-content" @default.
• W3172425208 hasAuthorship W3172425208A5014648528 @default.
• W3172425208 hasAuthorship W3172425208A5020523405 @default.
• W3172425208 hasAuthorship W3172425208A5025113992 @default.
• W3172425208 hasAuthorship W3172425208A5040003361 @default.
• W3172425208 hasAuthorship W3172425208A5069883194 @default.
• W3172425208 hasAuthorship W3172425208A5087256918 @default.
• W3172425208 hasConcept C107673813 @default.
• W3172425208 hasConcept C111919701 @default.
• W3172425208 hasConcept C118505674 @default.
• W3172425208 hasConcept C119857082 @default.
• W3172425208 hasConcept C153180895 @default.
• W3172425208 hasConcept C154945302 @default.
• W3172425208 hasConcept C165696696 @default.
• W3172425208 hasConcept C177769412 @default.
• W3172425208 hasConcept C2776214188 @default.
• W3172425208 hasConcept C2777212361 @default.
• W3172425208 hasConcept C2781238097 @default.
• W3172425208 hasConcept C38652104 @default.
• W3172425208 hasConcept C41008148 @default.
• W3172425208 hasConcept C81363708 @default.
• W3172425208 hasConceptScore W3172425208C107673813 @default.
• W3172425208 hasConceptScore W3172425208C111919701 @default.
• W3172425208 hasConceptScore W3172425208C118505674 @default.
• W3172425208 hasConceptScore W3172425208C119857082 @default.
• W3172425208 hasConceptScore W3172425208C153180895 @default.
• W3172425208 hasConceptScore W3172425208C154945302 @default.
• W3172425208 hasConceptScore W3172425208C165696696 @default.
• W3172425208 hasConceptScore W3172425208C177769412 @default.

• next