Matches in SemOpenAlex for { <https://semopenalex.org/work/W4321510871> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4321510871 abstract "The existing state-of-the-arts two-stage oriented object detectors have no significant improvement in the label assignment strategies, and the most widely-used one is the so-called Max IoU Assigner (MIA). In this paper, we first illustrate that MIA may cause matching conflicts in some cases, hinder the matching of ground-truth (GT) boxes with high-quality samples, which is extremely harmful to the training process. After that, we propose a Harmonized Label Assigner (HLA) for the oriented RPN, which can automatically harmonize the assignment priority of each GT box according to the corresponding number of candidate samples, solve the matching conflicts, and improve the detection accuracy of the two-stage oriented detectors. Finally, we implement the proposed HLA on Oriented R-CNN and conduct sufficient experiments on two public datasets (MAR20 and HRSC2016). Without tricks, our HLA significantly improves the detection accuracy of the detector to 83.97% mAP (on MAR20) and 90.42% mAP (on HRSC2016), respectively." @default.
- W4321510871 created "2023-02-24" @default.
- W4321510871 creator A5001838653 @default.
- W4321510871 creator A5004075655 @default.
- W4321510871 creator A5014157114 @default.
- W4321510871 creator A5047430724 @default.
- W4321510871 creator A5069913800 @default.
- W4321510871 date "2022-12-16" @default.
- W4321510871 modified "2023-09-25" @default.
- W4321510871 title "HLA: Harmonized Label Assigner for Two-stage Oriented Object Detection" @default.
- W4321510871 cites W2031489346 @default.
- W4321510871 cites W2194775991 @default.
- W4321510871 cites W2594177559 @default.
- W4321510871 cites W2963037989 @default.
- W4321510871 cites W2963150697 @default.
- W4321510871 cites W2963351448 @default.
- W4321510871 cites W2964979676 @default.
- W4321510871 cites W2982770724 @default.
- W4321510871 cites W2986357608 @default.
- W4321510871 cites W2991363140 @default.
- W4321510871 cites W3035396860 @default.
- W4321510871 cites W3107867277 @default.
- W4321510871 cites W3108849448 @default.
- W4321510871 cites W3136761610 @default.
- W4321510871 cites W3167308647 @default.
- W4321510871 cites W3170033848 @default.
- W4321510871 cites W3174873843 @default.
- W4321510871 cites W4214648418 @default.
- W4321510871 cites W4225125774 @default.
- W4321510871 cites W4312804579 @default.
- W4321510871 cites W639708223 @default.
- W4321510871 doi "https://doi.org/10.1109/icarce55724.2022.10046644" @default.
- W4321510871 hasPublicationYear "2022" @default.
- W4321510871 type Work @default.
- W4321510871 citedByCount "0" @default.
- W4321510871 crossrefType "proceedings-article" @default.
- W4321510871 hasAuthorship W4321510871A5001838653 @default.
- W4321510871 hasAuthorship W4321510871A5004075655 @default.
- W4321510871 hasAuthorship W4321510871A5014157114 @default.
- W4321510871 hasAuthorship W4321510871A5047430724 @default.
- W4321510871 hasAuthorship W4321510871A5069913800 @default.
- W4321510871 hasConcept C105795698 @default.
- W4321510871 hasConcept C111919701 @default.
- W4321510871 hasConcept C124101348 @default.
- W4321510871 hasConcept C146357865 @default.
- W4321510871 hasConcept C146849305 @default.
- W4321510871 hasConcept C151730666 @default.
- W4321510871 hasConcept C153180895 @default.
- W4321510871 hasConcept C154945302 @default.
- W4321510871 hasConcept C165064840 @default.
- W4321510871 hasConcept C2776151529 @default.
- W4321510871 hasConcept C2781238097 @default.
- W4321510871 hasConcept C33923547 @default.
- W4321510871 hasConcept C41008148 @default.
- W4321510871 hasConcept C76155785 @default.
- W4321510871 hasConcept C86803240 @default.
- W4321510871 hasConcept C94915269 @default.
- W4321510871 hasConcept C98045186 @default.
- W4321510871 hasConceptScore W4321510871C105795698 @default.
- W4321510871 hasConceptScore W4321510871C111919701 @default.
- W4321510871 hasConceptScore W4321510871C124101348 @default.
- W4321510871 hasConceptScore W4321510871C146357865 @default.
- W4321510871 hasConceptScore W4321510871C146849305 @default.
- W4321510871 hasConceptScore W4321510871C151730666 @default.
- W4321510871 hasConceptScore W4321510871C153180895 @default.
- W4321510871 hasConceptScore W4321510871C154945302 @default.
- W4321510871 hasConceptScore W4321510871C165064840 @default.
- W4321510871 hasConceptScore W4321510871C2776151529 @default.
- W4321510871 hasConceptScore W4321510871C2781238097 @default.
- W4321510871 hasConceptScore W4321510871C33923547 @default.
- W4321510871 hasConceptScore W4321510871C41008148 @default.
- W4321510871 hasConceptScore W4321510871C76155785 @default.
- W4321510871 hasConceptScore W4321510871C86803240 @default.
- W4321510871 hasConceptScore W4321510871C94915269 @default.
- W4321510871 hasConceptScore W4321510871C98045186 @default.
- W4321510871 hasLocation W43215108711 @default.
- W4321510871 hasOpenAccess W4321510871 @default.
- W4321510871 hasPrimaryLocation W43215108711 @default.
- W4321510871 hasRelatedWork W2019331124 @default.
- W4321510871 hasRelatedWork W2095705906 @default.
- W4321510871 hasRelatedWork W2801801420 @default.
- W4321510871 hasRelatedWork W2922421953 @default.
- W4321510871 hasRelatedWork W2970686063 @default.
- W4321510871 hasRelatedWork W2975200075 @default.
- W4321510871 hasRelatedWork W2998858795 @default.
- W4321510871 hasRelatedWork W3160733999 @default.
- W4321510871 hasRelatedWork W4210284706 @default.
- W4321510871 hasRelatedWork W4225418833 @default.
- W4321510871 isParatext "false" @default.
- W4321510871 isRetracted "false" @default.
- W4321510871 workType "article" @default.