FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3128660842> ?p ?o ?g. }

• W3128660842 endingPage "3460" @default.
• W3128660842 startingPage "3448" @default.
• W3128660842 abstract "Object detection has made enormous progress and has been widely used in many applications. However, it performs poorly when only limited training data is available for novel classes that the model has never seen before. Most existing approaches solve few-shot detection tasks implicitly without directly modeling the detectors for novel classes. In this article, we propose GenDet, a new meta-learning-based framework that can effectively generate object detectors for novel classes from few shots and, thus, conducts few-shot detection tasks explicitly. The detector generator is trained by numerous few-shot detection tasks sampled from base classes each with sufficient samples, and thus, it is expected to generalize well on novel classes. An adaptive pooling module is further introduced to suppress distracting samples and aggregate the detectors generated from multiple shots. Moreover, we propose to train a reference detector for each base class in the conventional way, with which to guide the training of the detector generator. The reference detectors and the detector generator can be trained simultaneously. Finally, the generated detectors of different classes are encouraged to be orthogonal to each other for better generalization. The proposed approach is extensively evaluated on the ImageNet, VOC, and COCO data sets under various few-shot detection settings, and it achieves new state-of-the-art results." @default.
• W3128660842 created "2021-02-15" @default.
• W3128660842 creator A5003676104 @default.
• W3128660842 creator A5009239895 @default.
• W3128660842 creator A5026184280 @default.
• W3128660842 creator A5037101612 @default.
• W3128660842 creator A5054299179 @default.
• W3128660842 creator A5057842914 @default.
• W3128660842 creator A5063370423 @default.
• W3128660842 creator A5079361172 @default.
• W3128660842 date "2022-08-01" @default.
• W3128660842 modified "2023-10-18" @default.
• W3128660842 title "GenDet: Meta Learning to Generate Detectors From Few Shots" @default.
• W3128660842 cites W1536680647 @default.
• W3128660842 cites W1922288304 @default.
• W3128660842 cites W2031489346 @default.
• W3128660842 cites W2108598243 @default.
• W3128660842 cites W2109255472 @default.
• W3128660842 cites W2115733720 @default.
• W3128660842 cites W2167460663 @default.
• W3128660842 cites W2183182206 @default.
• W3128660842 cites W2194321275 @default.
• W3128660842 cites W2194775991 @default.
• W3128660842 cites W2470394683 @default.
• W3128660842 cites W2519882289 @default.
• W3128660842 cites W2565639579 @default.
• W3128660842 cites W2601564443 @default.
• W3128660842 cites W2788159735 @default.
• W3128660842 cites W2788210750 @default.
• W3128660842 cites W2798836702 @default.
• W3128660842 cites W2799058067 @default.
• W3128660842 cites W2806070179 @default.
• W3128660842 cites W2886335102 @default.
• W3128660842 cites W2957105406 @default.
• W3128660842 cites W2962837320 @default.
• W3128660842 cites W2963037989 @default.
• W3128660842 cites W2963078860 @default.
• W3128660842 cites W2963299996 @default.
• W3128660842 cites W2963351448 @default.
• W3128660842 cites W2963559058 @default.
• W3128660842 cites W2963680240 @default.
• W3128660842 cites W2963845150 @default.
• W3128660842 cites W2964105864 @default.
• W3128660842 cites W2982770724 @default.
• W3128660842 cites W2983156430 @default.
• W3128660842 cites W2995360785 @default.
• W3128660842 cites W2996583130 @default.
• W3128660842 cites W2996976377 @default.
• W3128660842 cites W3009213340 @default.
• W3128660842 cites W3035316790 @default.
• W3128660842 cites W3099206234 @default.
• W3128660842 cites W3099554308 @default.
• W3128660842 cites W639708223 @default.
• W3128660842 doi "https://doi.org/10.1109/tnnls.2021.3053005" @default.
• W3128660842 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33523819" @default.
• W3128660842 hasPublicationYear "2022" @default.
• W3128660842 type Work @default.
• W3128660842 sameAs 3128660842 @default.
• W3128660842 citedByCount "8" @default.
• W3128660842 countsByYear W31286608422020 @default.
• W3128660842 countsByYear W31286608422021 @default.
• W3128660842 countsByYear W31286608422022 @default.
• W3128660842 countsByYear W31286608422023 @default.
• W3128660842 crossrefType "journal-article" @default.
• W3128660842 hasAuthorship W3128660842A5003676104 @default.
• W3128660842 hasAuthorship W3128660842A5009239895 @default.
• W3128660842 hasAuthorship W3128660842A5026184280 @default.
• W3128660842 hasAuthorship W3128660842A5037101612 @default.
• W3128660842 hasAuthorship W3128660842A5054299179 @default.
• W3128660842 hasAuthorship W3128660842A5057842914 @default.
• W3128660842 hasAuthorship W3128660842A5063370423 @default.
• W3128660842 hasAuthorship W3128660842A5079361172 @default.
• W3128660842 hasBestOaLocation W31286608422 @default.
• W3128660842 hasConcept C119857082 @default.
• W3128660842 hasConcept C120665830 @default.
• W3128660842 hasConcept C121332964 @default.
• W3128660842 hasConcept C134306372 @default.
• W3128660842 hasConcept C153180895 @default.
• W3128660842 hasConcept C154945302 @default.
• W3128660842 hasConcept C163258240 @default.
• W3128660842 hasConcept C177148314 @default.
• W3128660842 hasConcept C178790620 @default.
• W3128660842 hasConcept C185592680 @default.
• W3128660842 hasConcept C2776151529 @default.
• W3128660842 hasConcept C2778344882 @default.
• W3128660842 hasConcept C2780992000 @default.
• W3128660842 hasConcept C3019835501 @default.
• W3128660842 hasConcept C31972630 @default.
• W3128660842 hasConcept C33923547 @default.
• W3128660842 hasConcept C41008148 @default.
• W3128660842 hasConcept C62520636 @default.
• W3128660842 hasConcept C70437156 @default.
• W3128660842 hasConcept C76155785 @default.
• W3128660842 hasConcept C94915269 @default.
• W3128660842 hasConceptScore W3128660842C119857082 @default.
• W3128660842 hasConceptScore W3128660842C120665830 @default.
• W3128660842 hasConceptScore W3128660842C121332964 @default.
• W3128660842 hasConceptScore W3128660842C134306372 @default.

• next