FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2994648970> ?p ?o ?g. }

• W2994648970 abstract "The main requisite for fine-grained recognition task is to focus on subtle discriminative details that make the subordinate classes different from each other. We note that existing methods implicitly address this requirement and leave it to a data-driven pipeline to figure out what makes a subordinate class different from the others. This results in two major limitations: First, the network focuses on the most obvious distinctions between classes and overlooks more subtle inter-class variations. Second, the chance of misclassifying a given sample in any of the negative classes is considered equal, while in fact, confusions generally occur among only the most similar classes. Here, we propose to explicitly force the network to find the subtle differences among closely related classes. In this pursuit, we introduce two key novelties that can be easily plugged into existing end-to-end deep learning pipelines. On one hand, we introduce diversification block which masks the most salient features for an input to force the network to use more subtle cues for its correct classification. Concurrently, we introduce a gradient-boosting loss function that focuses only on the confusing classes for each sample and therefore moves swiftly along the direction on the loss surface that seeks to resolve these ambiguities. The synergy between these two blocks helps the network to learn more effective feature representations. Comprehensive experiments are performed on five challenging datasets. Our approach outperforms existing methods using similar experimental setting on all five datasets." @default.
• W2994648970 created "2019-12-26" @default.
• W2994648970 creator A5000300751 @default.
• W2994648970 creator A5009362997 @default.
• W2994648970 creator A5010215983 @default.
• W2994648970 creator A5079112987 @default.
• W2994648970 creator A5082634513 @default.
• W2994648970 date "2019-12-14" @default.
• W2994648970 modified "2023-09-23" @default.
• W2994648970 title "Fine-grained Recognition: Accounting for Subtle Differences between Similar Classes" @default.
• W2994648970 cites W1797268635 @default.
• W2994648970 cites W1846799578 @default.
• W2994648970 cites W1861492603 @default.
• W2994648970 cites W2104657103 @default.
• W2994648970 cites W2117539524 @default.
• W2994648970 cites W2118696714 @default.
• W2994648970 cites W2138011018 @default.
• W2994648970 cites W2183341477 @default.
• W2994648970 cites W2194775991 @default.
• W2994648970 cites W2295107390 @default.
• W2994648970 cites W2520774990 @default.
• W2994648970 cites W2566365295 @default.
• W2994648970 cites W2604195031 @default.
• W2994648970 cites W2737725206 @default.
• W2994648970 cites W2797977484 @default.
• W2994648970 cites W2798365843 @default.
• W2994648970 cites W2807931652 @default.
• W2994648970 cites W2883888092 @default.
• W2994648970 cites W2891951760 @default.
• W2994648970 cites W2892035828 @default.
• W2994648970 cites W2899771611 @default.
• W2994648970 cites W2961018736 @default.
• W2994648970 cites W2962761264 @default.
• W2994648970 cites W2963045696 @default.
• W2994648970 cites W2963066927 @default.
• W2994648970 cites W2963069818 @default.
• W2994648970 cites W2963150697 @default.
• W2994648970 cites W2963166243 @default.
• W2994648970 cites W2963190516 @default.
• W2994648970 cites W2963393555 @default.
• W2994648970 cites W2963407932 @default.
• W2994648970 cites W2964036919 @default.
• W2994648970 cites W2964176323 @default.
• W2994648970 doi "https://doi.org/10.48550/arxiv.1912.06842" @default.
• W2994648970 hasPublicationYear "2019" @default.
• W2994648970 type Work @default.
• W2994648970 sameAs 2994648970 @default.
• W2994648970 citedByCount "1" @default.
• W2994648970 countsByYear W29946489702020 @default.
• W2994648970 crossrefType "posted-content" @default.
• W2994648970 hasAuthorship W2994648970A5000300751 @default.
• W2994648970 hasAuthorship W2994648970A5009362997 @default.
• W2994648970 hasAuthorship W2994648970A5010215983 @default.
• W2994648970 hasAuthorship W2994648970A5079112987 @default.
• W2994648970 hasAuthorship W2994648970A5082634513 @default.
• W2994648970 hasBestOaLocation W29946489701 @default.
• W2994648970 hasConcept C119857082 @default.
• W2994648970 hasConcept C120665830 @default.
• W2994648970 hasConcept C121332964 @default.
• W2994648970 hasConcept C138885662 @default.
• W2994648970 hasConcept C14036430 @default.
• W2994648970 hasConcept C154945302 @default.
• W2994648970 hasConcept C192209626 @default.
• W2994648970 hasConcept C199360897 @default.
• W2994648970 hasConcept C2776401178 @default.
• W2994648970 hasConcept C2777212361 @default.
• W2994648970 hasConcept C2780719617 @default.
• W2994648970 hasConcept C41008148 @default.
• W2994648970 hasConcept C41895202 @default.
• W2994648970 hasConcept C43521106 @default.
• W2994648970 hasConcept C46686674 @default.
• W2994648970 hasConcept C78458016 @default.
• W2994648970 hasConcept C86803240 @default.
• W2994648970 hasConcept C97931131 @default.
• W2994648970 hasConceptScore W2994648970C119857082 @default.
• W2994648970 hasConceptScore W2994648970C120665830 @default.
• W2994648970 hasConceptScore W2994648970C121332964 @default.
• W2994648970 hasConceptScore W2994648970C138885662 @default.
• W2994648970 hasConceptScore W2994648970C14036430 @default.
• W2994648970 hasConceptScore W2994648970C154945302 @default.
• W2994648970 hasConceptScore W2994648970C192209626 @default.
• W2994648970 hasConceptScore W2994648970C199360897 @default.
• W2994648970 hasConceptScore W2994648970C2776401178 @default.
• W2994648970 hasConceptScore W2994648970C2777212361 @default.
• W2994648970 hasConceptScore W2994648970C2780719617 @default.
• W2994648970 hasConceptScore W2994648970C41008148 @default.
• W2994648970 hasConceptScore W2994648970C41895202 @default.
• W2994648970 hasConceptScore W2994648970C43521106 @default.
• W2994648970 hasConceptScore W2994648970C46686674 @default.
• W2994648970 hasConceptScore W2994648970C78458016 @default.
• W2994648970 hasConceptScore W2994648970C86803240 @default.
• W2994648970 hasConceptScore W2994648970C97931131 @default.
• W2994648970 hasLocation W29946489701 @default.
• W2994648970 hasLocation W29946489702 @default.
• W2994648970 hasOpenAccess W2994648970 @default.
• W2994648970 hasPrimaryLocation W29946489701 @default.
• W2994648970 hasRelatedWork W1987859285 @default.
• W2994648970 hasRelatedWork W1996541855 @default.
• W2994648970 hasRelatedWork W2026121273 @default.
• W2994648970 hasRelatedWork W2134990190 @default.

• next