FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3013755596> ?p ?o ?g. }

• W3013755596 abstract "Neural Architecture Search (NAS) has achieved great success in image classification task. Some recent works have managed to explore the automatic design of efficient backbone or feature fusion layer for object detection. However, these methods focus on searching only one certain component of object detector while leaving others manually designed. We identify the inconsistency between searched component and manually designed ones would withhold the detector of stronger performance. To this end, we propose a hierarchical trinity search framework to simultaneously discover efficient architectures for all components (i.e. backbone, neck, and head) of object detector in an end-to-end manner. In addition, we empirically reveal that different parts of the detector prefer different operators. Motivated by this, we employ a novel scheme to automatically screen different sub search spaces for different components so as to perform the end-to-end search for each component on the corresponding sub search space efficiently. Without bells and whistles, our searched architecture, namely Hit-Detector, achieves 41.4% mAP on COCO minival set with 27M parameters. Our implementation is available at https://github.com/ggjy/HitDet.pytorch." @default.
• W3013755596 created "2020-04-03" @default.
• W3013755596 creator A5000095574 @default.
• W3013755596 creator A5001529504 @default.
• W3013755596 creator A5006817088 @default.
• W3013755596 creator A5016185815 @default.
• W3013755596 creator A5043670420 @default.
• W3013755596 creator A5056485902 @default.
• W3013755596 creator A5062581018 @default.
• W3013755596 creator A5089966579 @default.
• W3013755596 date "2020-03-26" @default.
• W3013755596 modified "2023-09-22" @default.
• W3013755596 title "Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection" @default.
• W3013755596 cites W1536680647 @default.
• W3013755596 cites W1861492603 @default.
• W3013755596 cites W2088049833 @default.
• W3013755596 cites W2102605133 @default.
• W3013755596 cites W2194775991 @default.
• W3013755596 cites W2288122362 @default.
• W3013755596 cites W2407521645 @default.
• W3013755596 cites W2549139847 @default.
• W3013755596 cites W2553303224 @default.
• W3013755596 cites W2557728737 @default.
• W3013755596 cites W2565639579 @default.
• W3013755596 cites W2570343428 @default.
• W3013755596 cites W2579985080 @default.
• W3013755596 cites W2593744649 @default.
• W3013755596 cites W2599765304 @default.
• W3013755596 cites W2601564443 @default.
• W3013755596 cites W2613718673 @default.
• W3013755596 cites W2793693263 @default.
• W3013755596 cites W2796438033 @default.
• W3013755596 cites W2797527871 @default.
• W3013755596 cites W2810075754 @default.
• W3013755596 cites W2810857565 @default.
• W3013755596 cites W2883780447 @default.
• W3013755596 cites W2899771611 @default.
• W3013755596 cites W2905672847 @default.
• W3013755596 cites W2913691358 @default.
• W3013755596 cites W2925303509 @default.
• W3013755596 cites W2936599103 @default.
• W3013755596 cites W2938756873 @default.
• W3013755596 cites W2942263598 @default.
• W3013755596 cites W2946706208 @default.
• W3013755596 cites W2949264490 @default.
• W3013755596 cites W2949858991 @default.
• W3013755596 cites W2951886768 @default.
• W3013755596 cites W2963037989 @default.
• W3013755596 cites W2963351448 @default.
• W3013755596 cites W2963536136 @default.
• W3013755596 cites W2963778169 @default.
• W3013755596 cites W2963786238 @default.
• W3013755596 cites W2963821229 @default.
• W3013755596 cites W2963918968 @default.
• W3013755596 cites W2964080601 @default.
• W3013755596 cites W2964081807 @default.
• W3013755596 cites W2964241181 @default.
• W3013755596 cites W2964259004 @default.
• W3013755596 cites W2964444661 @default.
• W3013755596 cites W2965658867 @default.
• W3013755596 cites W2968333315 @default.
• W3013755596 cites W2970867386 @default.
• W3013755596 cites W2985228975 @default.
• W3013755596 cites W2988452521 @default.
• W3013755596 cites W3012573144 @default.
• W3013755596 cites W3106250896 @default.
• W3013755596 doi "https://doi.org/10.48550/arxiv.2003.11818" @default.
• W3013755596 hasPublicationYear "2020" @default.
• W3013755596 type Work @default.
• W3013755596 sameAs 3013755596 @default.
• W3013755596 citedByCount "5" @default.
• W3013755596 countsByYear W30137555962020 @default.
• W3013755596 countsByYear W30137555962021 @default.
• W3013755596 crossrefType "posted-content" @default.
• W3013755596 hasAuthorship W3013755596A5000095574 @default.
• W3013755596 hasAuthorship W3013755596A5001529504 @default.
• W3013755596 hasAuthorship W3013755596A5006817088 @default.
• W3013755596 hasAuthorship W3013755596A5016185815 @default.
• W3013755596 hasAuthorship W3013755596A5043670420 @default.
• W3013755596 hasAuthorship W3013755596A5056485902 @default.
• W3013755596 hasAuthorship W3013755596A5062581018 @default.
• W3013755596 hasAuthorship W3013755596A5089966579 @default.
• W3013755596 hasBestOaLocation W30137555961 @default.
• W3013755596 hasConcept C120665830 @default.
• W3013755596 hasConcept C121332964 @default.
• W3013755596 hasConcept C123657996 @default.
• W3013755596 hasConcept C127413603 @default.
• W3013755596 hasConcept C134306372 @default.
• W3013755596 hasConcept C138885662 @default.
• W3013755596 hasConcept C142362112 @default.
• W3013755596 hasConcept C153180895 @default.
• W3013755596 hasConcept C153349607 @default.
• W3013755596 hasConcept C154945302 @default.
• W3013755596 hasConcept C168167062 @default.
• W3013755596 hasConcept C177264268 @default.
• W3013755596 hasConcept C192209626 @default.
• W3013755596 hasConcept C199360897 @default.
• W3013755596 hasConcept C201995342 @default.
• W3013755596 hasConcept C2776151529 @default.
• W3013755596 hasConcept C2776401178 @default.

• next