FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2962798895> ?p ?o ?g. }

• W2962798895 abstract "Existing fine-grained visual categorization methods often suffer from three challenges: lack of training data, large number of fine-grained categories, and high intraclass vs. low inter-class variance. In this work we propose a generic iterative framework for fine-grained categorization and dataset bootstrapping that handles these three challenges. Using deep metric learning with humans in the loop, we learn a low dimensional feature embedding with anchor points on manifolds for each category. These anchor points capture intra-class variances and remain discriminative between classes. In each round, images with high confidence scores from our model are sent to humans for labeling. By comparing with exemplar images, labelers mark each candidate image as either a true positive or a false positive. True positives are added into our current dataset and false positives are regarded as hard negatives for our metric learning model. Then the model is retrained with an expanded dataset and hard negatives for the next round. To demonstrate the effectiveness of the proposed framework, we bootstrap a fine-grained flower dataset with 620 categories from Instagram images. The proposed deep metric learning scheme is evaluated on both our dataset and the CUB-200-2001 Birds dataset. Experimental evaluations show significant performance gain using dataset bootstrapping and demonstrate state-of-the-art results achieved by the proposed deep metric learning methods." @default.
• W2962798895 created "2019-07-30" @default.
• W2962798895 creator A5018609918 @default.
• W2962798895 creator A5041864624 @default.
• W2962798895 creator A5054137460 @default.
• W2962798895 creator A5088559134 @default.
• W2962798895 date "2016-06-01" @default.
• W2962798895 modified "2023-10-11" @default.
• W2962798895 title "Fine-Grained Categorization and Dataset Bootstrapping Using Deep Metric Learning with Humans in the Loop" @default.
• W2962798895 cites W1506491340 @default.
• W2962798895 cites W1898560071 @default.
• W2962798895 cites W1927052826 @default.
• W2962798895 cites W1939575207 @default.
• W2962798895 cites W1946093182 @default.
• W2962798895 cites W1946609740 @default.
• W2962798895 cites W1954152232 @default.
• W2962798895 cites W1958236864 @default.
• W2962798895 cites W1975517671 @default.
• W2962798895 cites W1995543189 @default.
• W2962798895 cites W2014102544 @default.
• W2962798895 cites W2021354639 @default.
• W2962798895 cites W2102605133 @default.
• W2962798895 cites W2104657103 @default.
• W2962798895 cites W2110015572 @default.
• W2962798895 cites W2118696714 @default.
• W2962798895 cites W2129492014 @default.
• W2962798895 cites W2135706578 @default.
• W2962798895 cites W2138011018 @default.
• W2962798895 cites W2138621090 @default.
• W2962798895 cites W2145287260 @default.
• W2962798895 cites W2150856297 @default.
• W2962798895 cites W2157364932 @default.
• W2962798895 cites W2157383331 @default.
• W2962798895 cites W2168644216 @default.
• W2962798895 cites W2221507685 @default.
• W2962798895 cites W2533598788 @default.
• W2962798895 cites W2964176323 @default.
• W2962798895 cites W3099206234 @default.
• W2962798895 doi "https://doi.org/10.1109/cvpr.2016.130" @default.
• W2962798895 hasPublicationYear "2016" @default.
• W2962798895 type Work @default.
• W2962798895 sameAs 2962798895 @default.
• W2962798895 citedByCount "170" @default.
• W2962798895 countsByYear W29627988952016 @default.
• W2962798895 countsByYear W29627988952017 @default.
• W2962798895 countsByYear W29627988952018 @default.
• W2962798895 countsByYear W29627988952019 @default.
• W2962798895 countsByYear W29627988952020 @default.
• W2962798895 countsByYear W29627988952021 @default.
• W2962798895 countsByYear W29627988952022 @default.
• W2962798895 countsByYear W29627988952023 @default.
• W2962798895 crossrefType "proceedings-article" @default.
• W2962798895 hasAuthorship W2962798895A5018609918 @default.
• W2962798895 hasAuthorship W2962798895A5041864624 @default.
• W2962798895 hasAuthorship W2962798895A5054137460 @default.
• W2962798895 hasAuthorship W2962798895A5088559134 @default.
• W2962798895 hasBestOaLocation W29627988952 @default.
• W2962798895 hasConcept C108583219 @default.
• W2962798895 hasConcept C112789634 @default.
• W2962798895 hasConcept C119857082 @default.
• W2962798895 hasConcept C121955636 @default.
• W2962798895 hasConcept C138885662 @default.
• W2962798895 hasConcept C144133560 @default.
• W2962798895 hasConcept C149782125 @default.
• W2962798895 hasConcept C153180895 @default.
• W2962798895 hasConcept C154945302 @default.
• W2962798895 hasConcept C162324750 @default.
• W2962798895 hasConcept C176217482 @default.
• W2962798895 hasConcept C196083921 @default.
• W2962798895 hasConcept C207609745 @default.
• W2962798895 hasConcept C21547014 @default.
• W2962798895 hasConcept C22019652 @default.
• W2962798895 hasConcept C2776401178 @default.
• W2962798895 hasConcept C2777212361 @default.
• W2962798895 hasConcept C33923547 @default.
• W2962798895 hasConcept C41008148 @default.
• W2962798895 hasConcept C41608201 @default.
• W2962798895 hasConcept C41895202 @default.
• W2962798895 hasConcept C50644808 @default.
• W2962798895 hasConcept C64869954 @default.
• W2962798895 hasConcept C94124525 @default.
• W2962798895 hasConcept C97931131 @default.
• W2962798895 hasConceptScore W2962798895C108583219 @default.
• W2962798895 hasConceptScore W2962798895C112789634 @default.
• W2962798895 hasConceptScore W2962798895C119857082 @default.
• W2962798895 hasConceptScore W2962798895C121955636 @default.
• W2962798895 hasConceptScore W2962798895C138885662 @default.
• W2962798895 hasConceptScore W2962798895C144133560 @default.
• W2962798895 hasConceptScore W2962798895C149782125 @default.
• W2962798895 hasConceptScore W2962798895C153180895 @default.
• W2962798895 hasConceptScore W2962798895C154945302 @default.
• W2962798895 hasConceptScore W2962798895C162324750 @default.
• W2962798895 hasConceptScore W2962798895C176217482 @default.
• W2962798895 hasConceptScore W2962798895C196083921 @default.
• W2962798895 hasConceptScore W2962798895C207609745 @default.
• W2962798895 hasConceptScore W2962798895C21547014 @default.
• W2962798895 hasConceptScore W2962798895C22019652 @default.
• W2962798895 hasConceptScore W2962798895C2776401178 @default.
• W2962798895 hasConceptScore W2962798895C2777212361 @default.
• W2962798895 hasConceptScore W2962798895C33923547 @default.

• next