Matches in SemOpenAlex for { <https://semopenalex.org/work/W4319300007> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W4319300007 abstract "There is evidence that transformers offer state-of-the-art recognition performance on tasks involving overhead imagery (e.g., satellite imagery). However, it is difficult to make unbiased empirical comparisons between competing deep learning models, making it unclear whether, and to what extent, transformer-based models are beneficial. In this paper we systematically compare the impact of adding transformer structures into state-of-the-art segmentation models for overhead imagery. Each model is given a similar budget of free parameters, and their hyperparameters are optimized using Bayesian Optimization with a fixed quantity of data and computation time. We conduct our experiments with a large and diverse dataset comprising two large public benchmarks: Inria and DeepGlobe. We perform additional ablation studies to explore the impact of specific transformer-based modeling choices. Our results suggest that transformers provide consistent, but modest, performance improvements. We only observe this advantage however in hybrid models that combine convolutional and transformer-based structures, while fully transformer-based models achieve relatively poor performance." @default.
- W4319300007 created "2023-02-07" @default.
- W4319300007 creator A5004578492 @default.
- W4319300007 creator A5007239332 @default.
- W4319300007 creator A5010537775 @default.
- W4319300007 creator A5054438743 @default.
- W4319300007 creator A5081327514 @default.
- W4319300007 date "2023-01-01" @default.
- W4319300007 modified "2023-09-29" @default.
- W4319300007 title "Transformers For Recognition In Overhead Imagery: A Reality Check" @default.
- W4319300007 cites W1510052597 @default.
- W4319300007 cites W2609402060 @default.
- W4319300007 cites W2787225861 @default.
- W4319300007 cites W2804199516 @default.
- W4319300007 cites W2804962028 @default.
- W4319300007 cites W2806581075 @default.
- W4319300007 cites W2884436604 @default.
- W4319300007 cites W2907750714 @default.
- W4319300007 cites W2922544158 @default.
- W4319300007 cites W2962766617 @default.
- W4319300007 cites W2962914239 @default.
- W4319300007 cites W2964227007 @default.
- W4319300007 cites W3011515952 @default.
- W4319300007 cites W3015788359 @default.
- W4319300007 cites W3048802707 @default.
- W4319300007 cites W3100865231 @default.
- W4319300007 cites W3109684201 @default.
- W4319300007 cites W3115190122 @default.
- W4319300007 cites W3138516171 @default.
- W4319300007 cites W3161120562 @default.
- W4319300007 cites W3161790566 @default.
- W4319300007 cites W3163631062 @default.
- W4319300007 cites W3171660447 @default.
- W4319300007 cites W3204255739 @default.
- W4319300007 cites W3204614423 @default.
- W4319300007 cites W3205020875 @default.
- W4319300007 cites W3209987414 @default.
- W4319300007 cites W3211329537 @default.
- W4319300007 cites W4205138939 @default.
- W4319300007 cites W4214760051 @default.
- W4319300007 cites W4283450732 @default.
- W4319300007 cites W4312919733 @default.
- W4319300007 doi "https://doi.org/10.1109/wacv56688.2023.00377" @default.
- W4319300007 hasPublicationYear "2023" @default.
- W4319300007 type Work @default.
- W4319300007 citedByCount "0" @default.
- W4319300007 crossrefType "proceedings-article" @default.
- W4319300007 hasAuthorship W4319300007A5004578492 @default.
- W4319300007 hasAuthorship W4319300007A5007239332 @default.
- W4319300007 hasAuthorship W4319300007A5010537775 @default.
- W4319300007 hasAuthorship W4319300007A5054438743 @default.
- W4319300007 hasAuthorship W4319300007A5081327514 @default.
- W4319300007 hasBestOaLocation W43193000072 @default.
- W4319300007 hasConcept C11413529 @default.
- W4319300007 hasConcept C119599485 @default.
- W4319300007 hasConcept C119857082 @default.
- W4319300007 hasConcept C127413603 @default.
- W4319300007 hasConcept C154945302 @default.
- W4319300007 hasConcept C165801399 @default.
- W4319300007 hasConcept C41008148 @default.
- W4319300007 hasConcept C45374587 @default.
- W4319300007 hasConcept C66322947 @default.
- W4319300007 hasConcept C8642999 @default.
- W4319300007 hasConcept C89600930 @default.
- W4319300007 hasConceptScore W4319300007C11413529 @default.
- W4319300007 hasConceptScore W4319300007C119599485 @default.
- W4319300007 hasConceptScore W4319300007C119857082 @default.
- W4319300007 hasConceptScore W4319300007C127413603 @default.
- W4319300007 hasConceptScore W4319300007C154945302 @default.
- W4319300007 hasConceptScore W4319300007C165801399 @default.
- W4319300007 hasConceptScore W4319300007C41008148 @default.
- W4319300007 hasConceptScore W4319300007C45374587 @default.
- W4319300007 hasConceptScore W4319300007C66322947 @default.
- W4319300007 hasConceptScore W4319300007C8642999 @default.
- W4319300007 hasConceptScore W4319300007C89600930 @default.
- W4319300007 hasFunder F4320306151 @default.
- W4319300007 hasLocation W43193000071 @default.
- W4319300007 hasLocation W43193000072 @default.
- W4319300007 hasOpenAccess W4319300007 @default.
- W4319300007 hasPrimaryLocation W43193000071 @default.
- W4319300007 hasRelatedWork W3199608561 @default.
- W4319300007 hasRelatedWork W4210794429 @default.
- W4319300007 hasRelatedWork W4223456145 @default.
- W4319300007 hasRelatedWork W4280535922 @default.
- W4319300007 hasRelatedWork W4283697347 @default.
- W4319300007 hasRelatedWork W4295309597 @default.
- W4319300007 hasRelatedWork W4295681619 @default.
- W4319300007 hasRelatedWork W4304128395 @default.
- W4319300007 hasRelatedWork W4307195028 @default.
- W4319300007 hasRelatedWork W4309113015 @default.
- W4319300007 isParatext "false" @default.
- W4319300007 isRetracted "false" @default.
- W4319300007 workType "article" @default.