FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3134020009> ?p ?o ?g. }

• W3134020009 abstract "To safely deploy autonomous vehicles, onboard perception systems must work reliably at high accuracy across a diverse set of environments and geographies. One of the most common techniques to improve the efficacy of such systems in new domains involves collecting large labeled datasets, but such datasets can be extremely costly to obtain, especially if each new deployment geography requires additional data with expensive 3D bounding box annotations. We demonstrate that pseudo-labeling for 3D object detection is an effective way to exploit less expensive and more widely available unlabeled data, and can lead to performance gains across various architectures, data augmentation strategies, and sizes of the labeled dataset. Overall, we show that better teacher models lead to better student models, and that we can distill expensive teachers into efficient, simple students. Specifically, we demonstrate that pseudo-label-trained student models can outperform supervised models trained on 3-10 times the amount of labeled examples. Using PointPillars [24], a two-year-old architecture, as our student model, we are able to achieve state of the art accuracy simply by leveraging large quantities of pseudo-labeled data. Lastly, we show that these student models generalize better than supervised models to a new domain in which we only have unlabeled data, making pseudo-label training an effective form of unsupervised domain adaptation." @default.
• W3134020009 created "2021-03-15" @default.
• W3134020009 creator A5004650249 @default.
• W3134020009 creator A5008334918 @default.
• W3134020009 creator A5008894717 @default.
• W3134020009 creator A5032356827 @default.
• W3134020009 creator A5045825594 @default.
• W3134020009 creator A5056830118 @default.
• W3134020009 creator A5079961895 @default.
• W3134020009 date "2021-03-02" @default.
• W3134020009 modified "2023-09-27" @default.
• W3134020009 title "Pseudo-labeling for Scalable 3D Object Detection." @default.
• W3134020009 cites W1522301498 @default.
• W3134020009 cites W1731081199 @default.
• W3134020009 cites W1821462560 @default.
• W3134020009 cites W1923184257 @default.
• W3134020009 cites W1975165783 @default.
• W3134020009 cites W2018573225 @default.
• W3134020009 cites W2079057609 @default.
• W3134020009 cites W2115579991 @default.
• W3134020009 cites W2134135198 @default.
• W3134020009 cites W2154844948 @default.
• W3134020009 cites W2221898772 @default.
• W3134020009 cites W2592691248 @default.
• W3134020009 cites W2773446523 @default.
• W3134020009 cites W2798930779 @default.
• W3134020009 cites W2798965597 @default.
• W3134020009 cites W2897529137 @default.
• W3134020009 cites W2899302124 @default.
• W3134020009 cites W2943152387 @default.
• W3134020009 cites W2951970475 @default.
• W3134020009 cites W2962900737 @default.
• W3134020009 cites W2963060032 @default.
• W3134020009 cites W2963435192 @default.
• W3134020009 cites W2963709863 @default.
• W3134020009 cites W2963727135 @default.
• W3134020009 cites W2964153729 @default.
• W3134020009 cites W2968296999 @default.
• W3134020009 cites W2970259716 @default.
• W3134020009 cites W2978426779 @default.
• W3134020009 cites W2981581281 @default.
• W3134020009 cites W2989700832 @default.
• W3134020009 cites W2995681297 @default.
• W3134020009 cites W3001197829 @default.
• W3134020009 cites W3015033898 @default.
• W3134020009 cites W3015522062 @default.
• W3134020009 cites W3021542222 @default.
• W3134020009 cites W3026041220 @default.
• W3134020009 cites W3027513153 @default.
• W3134020009 cites W3028227066 @default.
• W3134020009 cites W3031752193 @default.
• W3134020009 cites W3034602892 @default.
• W3134020009 cites W3034975685 @default.
• W3134020009 cites W3035003500 @default.
• W3134020009 cites W3035160371 @default.
• W3134020009 cites W3035172746 @default.
• W3134020009 cites W3035363555 @default.
• W3134020009 cites W3035574168 @default.
• W3134020009 cites W3036853234 @default.
• W3134020009 cites W3037058446 @default.
• W3134020009 cites W3037262931 @default.
• W3134020009 cites W3041610550 @default.
• W3134020009 cites W3088900620 @default.
• W3134020009 cites W3108381027 @default.
• W3134020009 cites W3114286648 @default.
• W3134020009 cites W3122151369 @default.
• W3134020009 cites W3208394352 @default.
• W3134020009 hasPublicationYear "2021" @default.
• W3134020009 type Work @default.
• W3134020009 sameAs 3134020009 @default.
• W3134020009 citedByCount "5" @default.
• W3134020009 countsByYear W31340200092021 @default.
• W3134020009 crossrefType "posted-content" @default.
• W3134020009 hasAuthorship W3134020009A5004650249 @default.
• W3134020009 hasAuthorship W3134020009A5008334918 @default.
• W3134020009 hasAuthorship W3134020009A5008894717 @default.
• W3134020009 hasAuthorship W3134020009A5032356827 @default.
• W3134020009 hasAuthorship W3134020009A5045825594 @default.
• W3134020009 hasAuthorship W3134020009A5056830118 @default.
• W3134020009 hasAuthorship W3134020009A5079961895 @default.
• W3134020009 hasConcept C105339364 @default.
• W3134020009 hasConcept C111919701 @default.
• W3134020009 hasConcept C115961682 @default.
• W3134020009 hasConcept C119857082 @default.
• W3134020009 hasConcept C120665830 @default.
• W3134020009 hasConcept C121332964 @default.
• W3134020009 hasConcept C124101348 @default.
• W3134020009 hasConcept C134306372 @default.
• W3134020009 hasConcept C139807058 @default.
• W3134020009 hasConcept C147037132 @default.
• W3134020009 hasConcept C153180895 @default.
• W3134020009 hasConcept C154945302 @default.
• W3134020009 hasConcept C165696696 @default.
• W3134020009 hasConcept C177264268 @default.
• W3134020009 hasConcept C199360897 @default.
• W3134020009 hasConcept C2776145971 @default.
• W3134020009 hasConcept C2776151529 @default.
• W3134020009 hasConcept C2776434776 @default.
• W3134020009 hasConcept C2781238097 @default.
• W3134020009 hasConcept C33923547 @default.

• next