FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2959784504> ?p ?o ?g. }

• W2959784504 endingPage "159" @default.
• W2959784504 startingPage "150" @default.
• W2959784504 abstract "Fine-grained image classification aims to distinguish subcategories belonging to the same basic-level category, such as 200 subcategories belonging to bird. It is a challenging problem in computer vision and multimedia field due to: attribute similarity (e.g. color and texture) among different subcategories and attribute variance (e.g. pose and viewpoint) in the same subcategory. Attribute similarity causes the difficulty to classify different subcategories even for human, while attribute variance causes the learned feature representations chaotic and confused. Naturally, classification can benefit from a hierarchy of subcategories: since going to a coarser granularity leverages high-level semantic features, while going to a finer granularity leverages discriminative and subtle features. Therefore, we propose an attribute hierarchy based multi-task learning (AHMTL) approach, and its main novelties are: (1) Attribute hierarchy: We reassign all images to multi-granularity subcategories automatically, namely coarse-grained, fine-grained and ultra-fine-grained subcategories. Similar fine-grained subcategories are reassigned to the same coarse-grained subcategory according to their attribute similarity, which pays more attention to the high-level semantic feature representations. Simultaneously, the same fine-grained subcategory but with different attributes are divided into different ultra-fine-grained subcategories, which can obtain more discriminative and subtle feature representations for attribute variance. (2) Multi-task learning: A multi-task learning framework is designed to effectively learn robust feature representations by jointly optimizing coarse-grained, fine-grained and ultra-fine-grained image classification tasks. These three level tasks learn coarse to fine feature representations, meaning high-level semantic to subtle features, which can regularize and boost each other to prevent overfitting, and the mutual promotion of them ensures feature representations more discriminative. Compared with more than 10 state-of-the-art methods on two widely-used CUB-200-2011 and Cars-196 datasets, our AHMTL approach achieves the best classification performance." @default.
• W2959784504 created "2019-07-23" @default.
• W2959784504 creator A5017511861 @default.
• W2959784504 creator A5047811387 @default.
• W2959784504 creator A5060500814 @default.
• W2959784504 date "2020-06-01" @default.
• W2959784504 modified "2023-10-04" @default.
• W2959784504 title "Attribute hierarchy based multi-task learning for fine-grained image classification" @default.
• W2959784504 cites W1967632700 @default.
• W2959784504 cites W2151103935 @default.
• W2959784504 cites W2194011657 @default.
• W2959784504 cites W2196975585 @default.
• W2959784504 cites W2247349754 @default.
• W2959784504 cites W2289708887 @default.
• W2959784504 cites W2342491128 @default.
• W2959784504 cites W2342880667 @default.
• W2959784504 cites W2515116636 @default.
• W2959784504 cites W2544405876 @default.
• W2959784504 cites W2581874211 @default.
• W2959784504 cites W2586934430 @default.
• W2959784504 cites W2605793178 @default.
• W2959784504 cites W2963090248 @default.
• W2959784504 cites W3124951096 @default.
• W2959784504 doi "https://doi.org/10.1016/j.neucom.2018.02.109" @default.
• W2959784504 hasPublicationYear "2020" @default.
• W2959784504 type Work @default.
• W2959784504 sameAs 2959784504 @default.
• W2959784504 citedByCount "24" @default.
• W2959784504 countsByYear W29597845042019 @default.
• W2959784504 countsByYear W29597845042021 @default.
• W2959784504 countsByYear W29597845042022 @default.
• W2959784504 countsByYear W29597845042023 @default.
• W2959784504 crossrefType "journal-article" @default.
• W2959784504 hasAuthorship W2959784504A5017511861 @default.
• W2959784504 hasAuthorship W2959784504A5047811387 @default.
• W2959784504 hasAuthorship W2959784504A5060500814 @default.
• W2959784504 hasConcept C103278499 @default.
• W2959784504 hasConcept C111919701 @default.
• W2959784504 hasConcept C115961682 @default.
• W2959784504 hasConcept C119857082 @default.
• W2959784504 hasConcept C121955636 @default.
• W2959784504 hasConcept C138885662 @default.
• W2959784504 hasConcept C144133560 @default.
• W2959784504 hasConcept C153180895 @default.
• W2959784504 hasConcept C154945302 @default.
• W2959784504 hasConcept C162324750 @default.
• W2959784504 hasConcept C177774035 @default.
• W2959784504 hasConcept C187736073 @default.
• W2959784504 hasConcept C196083921 @default.
• W2959784504 hasConcept C202444582 @default.
• W2959784504 hasConcept C2776401178 @default.
• W2959784504 hasConcept C2780451532 @default.
• W2959784504 hasConcept C2780617661 @default.
• W2959784504 hasConcept C31170391 @default.
• W2959784504 hasConcept C33923547 @default.
• W2959784504 hasConcept C34447519 @default.
• W2959784504 hasConcept C41008148 @default.
• W2959784504 hasConcept C41895202 @default.
• W2959784504 hasConcept C97931131 @default.
• W2959784504 hasConceptScore W2959784504C103278499 @default.
• W2959784504 hasConceptScore W2959784504C111919701 @default.
• W2959784504 hasConceptScore W2959784504C115961682 @default.
• W2959784504 hasConceptScore W2959784504C119857082 @default.
• W2959784504 hasConceptScore W2959784504C121955636 @default.
• W2959784504 hasConceptScore W2959784504C138885662 @default.
• W2959784504 hasConceptScore W2959784504C144133560 @default.
• W2959784504 hasConceptScore W2959784504C153180895 @default.
• W2959784504 hasConceptScore W2959784504C154945302 @default.
• W2959784504 hasConceptScore W2959784504C162324750 @default.
• W2959784504 hasConceptScore W2959784504C177774035 @default.
• W2959784504 hasConceptScore W2959784504C187736073 @default.
• W2959784504 hasConceptScore W2959784504C196083921 @default.
• W2959784504 hasConceptScore W2959784504C202444582 @default.
• W2959784504 hasConceptScore W2959784504C2776401178 @default.
• W2959784504 hasConceptScore W2959784504C2780451532 @default.
• W2959784504 hasConceptScore W2959784504C2780617661 @default.
• W2959784504 hasConceptScore W2959784504C31170391 @default.
• W2959784504 hasConceptScore W2959784504C33923547 @default.
• W2959784504 hasConceptScore W2959784504C34447519 @default.
• W2959784504 hasConceptScore W2959784504C41008148 @default.
• W2959784504 hasConceptScore W2959784504C41895202 @default.
• W2959784504 hasConceptScore W2959784504C97931131 @default.
• W2959784504 hasFunder F4320321001 @default.
• W2959784504 hasLocation W29597845041 @default.
• W2959784504 hasOpenAccess W2959784504 @default.
• W2959784504 hasPrimaryLocation W29597845041 @default.
• W2959784504 hasRelatedWork W1972656095 @default.
• W2959784504 hasRelatedWork W2024160000 @default.
• W2959784504 hasRelatedWork W2061273563 @default.
• W2959784504 hasRelatedWork W2285052147 @default.
• W2959784504 hasRelatedWork W2729514902 @default.
• W2959784504 hasRelatedWork W2743258233 @default.
• W2959784504 hasRelatedWork W2773500201 @default.
• W2959784504 hasRelatedWork W2970216048 @default.
• W2959784504 hasRelatedWork W2998168123 @default.
• W2959784504 hasRelatedWork W4287995534 @default.
• W2959784504 hasVolume "395" @default.
• W2959784504 isParatext "false" @default.

• next