FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3090448772> ?p ?o ?g. }

• W3090448772 abstract "Simple image rotations significantly reduce the accuracy of deep neural networks. Moreover, training with all possible rotations increases the data set, which also increases the training duration. In this work, we address trainable rotation invariant convolutions as well as the construction of nets, since fully connected layers can only be rotation invariant with a one-dimensional input. On the one hand, we show that our approach is rotationally invariant for different models and on different public data sets. We also discuss the influence of purely rotational invariant features on accuracy. The rotationally adaptive convolution models presented in this work are more computationally intensive than normal convolution models. Therefore, we also present a depth wise separable approach with radial convolution. Link to CUDA code this https URL" @default.
• W3090448772 created "2020-10-08" @default.
• W3090448772 creator A5008809634 @default.
• W3090448772 creator A5015458683 @default.
• W3090448772 date "2020-10-02" @default.
• W3090448772 modified "2023-09-24" @default.
• W3090448772 title "Rotated Ring, Radial and Depth Wise Separable Radial Convolutions" @default.
• W3090448772 cites W1522301498 @default.
• W3090448772 cites W1533861849 @default.
• W3090448772 cites W1538131130 @default.
• W3090448772 cites W1901129140 @default.
• W3090448772 cites W1903029394 @default.
• W3090448772 cites W1980287119 @default.
• W3090448772 cites W2037227137 @default.
• W3090448772 cites W2044738244 @default.
• W3090448772 cites W2115252128 @default.
• W3090448772 cites W2194775991 @default.
• W3090448772 cites W2279221249 @default.
• W3090448772 cites W2340427832 @default.
• W3090448772 cites W2519332860 @default.
• W3090448772 cites W2531836693 @default.
• W3090448772 cites W2569680626 @default.
• W3090448772 cites W2576915720 @default.
• W3090448772 cites W2594660495 @default.
• W3090448772 cites W2600633878 @default.
• W3090448772 cites W2601564443 @default.
• W3090448772 cites W2612445135 @default.
• W3090448772 cites W2613634806 @default.
• W3090448772 cites W2640329709 @default.
• W3090448772 cites W2647318755 @default.
• W3090448772 cites W2730444669 @default.
• W3090448772 cites W2767702016 @default.
• W3090448772 cites W2795783309 @default.
• W3090448772 cites W2796409754 @default.
• W3090448772 cites W2805466601 @default.
• W3090448772 cites W2806382857 @default.
• W3090448772 cites W2889190845 @default.
• W3090448772 cites W2899694408 @default.
• W3090448772 cites W2900427309 @default.
• W3090448772 cites W2914528607 @default.
• W3090448772 cites W2917262027 @default.
• W3090448772 cites W2941208326 @default.
• W3090448772 cites W2946948240 @default.
• W3090448772 cites W2947365893 @default.
• W3090448772 cites W2949117887 @default.
• W3090448772 cites W2960416371 @default.
• W3090448772 cites W2963299736 @default.
• W3090448772 cites W2963826370 @default.
• W3090448772 cites W2966661 @default.
• W3090448772 cites W2969498159 @default.
• W3090448772 cites W2997768619 @default.
• W3090448772 cites W2998661336 @default.
• W3090448772 cites W3005287825 @default.
• W3090448772 cites W3010721491 @default.
• W3090448772 cites W3028642772 @default.
• W3090448772 cites W3033290417 @default.
• W3090448772 cites W3034092125 @default.
• W3090448772 cites W3080155310 @default.
• W3090448772 cites W3100972044 @default.
• W3090448772 cites W3102779874 @default.
• W3090448772 cites W3118608800 @default.
• W3090448772 cites W3160145939 @default.
• W3090448772 cites W3162448371 @default.
• W3090448772 cites W603908379 @default.
• W3090448772 cites W2933099034 @default.
• W3090448772 hasPublicationYear "2020" @default.
• W3090448772 type Work @default.
• W3090448772 sameAs 3090448772 @default.
• W3090448772 citedByCount "4" @default.
• W3090448772 countsByYear W30904487722021 @default.
• W3090448772 countsByYear W30904487722023 @default.
• W3090448772 crossrefType "posted-content" @default.
• W3090448772 hasAuthorship W3090448772A5008809634 @default.
• W3090448772 hasAuthorship W3090448772A5015458683 @default.
• W3090448772 hasConcept C11413529 @default.
• W3090448772 hasConcept C134306372 @default.
• W3090448772 hasConcept C154945302 @default.
• W3090448772 hasConcept C190470478 @default.
• W3090448772 hasConcept C33923547 @default.
• W3090448772 hasConcept C37914503 @default.
• W3090448772 hasConcept C41008148 @default.
• W3090448772 hasConcept C45347329 @default.
• W3090448772 hasConcept C50644808 @default.
• W3090448772 hasConcept C70710897 @default.
• W3090448772 hasConcept C74050887 @default.
• W3090448772 hasConcept C81363708 @default.
• W3090448772 hasConceptScore W3090448772C11413529 @default.
• W3090448772 hasConceptScore W3090448772C134306372 @default.
• W3090448772 hasConceptScore W3090448772C154945302 @default.
• W3090448772 hasConceptScore W3090448772C190470478 @default.
• W3090448772 hasConceptScore W3090448772C33923547 @default.
• W3090448772 hasConceptScore W3090448772C37914503 @default.
• W3090448772 hasConceptScore W3090448772C41008148 @default.
• W3090448772 hasConceptScore W3090448772C45347329 @default.
• W3090448772 hasConceptScore W3090448772C50644808 @default.
• W3090448772 hasConceptScore W3090448772C70710897 @default.
• W3090448772 hasConceptScore W3090448772C74050887 @default.
• W3090448772 hasConceptScore W3090448772C81363708 @default.
• W3090448772 hasLocation W30904487721 @default.
• W3090448772 hasOpenAccess W3090448772 @default.

• next