Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312823189> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W4312823189 endingPage "1927" @default.
- W4312823189 startingPage "1916" @default.
- W4312823189 abstract "Product detection, which aims to localize products of interest in the advertising images, helps advance many potential E-commerce applications like product retrieval and recommendation. However, labeling a massive number of fine-grained product categories and accurate product boxes is costly and especially not practical since products are ever-changing on E-commerce websites. In this work, we step forward to train a fine-grained product detector solely supervised by the advertising captions, which are naturally available but often severely flawed and noisy. To reformulate the weakly supervised detection research into a real-world setting, we introduce a large-scale benchmark, named <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>CapProduct</i> , where more than 80,000 product image-caption pairs are collected from E-commerce websites. The fine-grained nature of products and noisy captions in <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>CapProduct</i> make it intractable to excavate valid category labels to train a weakly supervised object detector. To tackle this challenge, we propose a <bold xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>Co</b> llaborative <bold xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>P</b> seudo- <bold xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>L</b> abel <bold xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>H</b> armonization (CoPLH) framework that harmonizes self-mined pseudo labels via modeling the global co-occurrence relationships of products. We construct a collaborative co-occurrence graph based on all training samples to improve the reliability of caption-predicted pseudo-labels as well as benefit the self-training procedure in a weakly supervised setting. Extensive experiments on the <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>CapProduct</i> dataset demonstrate the effectiveness and the superiority of the proposed CoPLH over the state-of-the-art baselines." @default.
- W4312823189 created "2023-01-05" @default.
- W4312823189 creator A5005325306 @default.
- W4312823189 creator A5037966954 @default.
- W4312823189 creator A5043757808 @default.
- W4312823189 creator A5047878798 @default.
- W4312823189 creator A5048961496 @default.
- W4312823189 creator A5049388503 @default.
- W4312823189 creator A5087043856 @default.
- W4312823189 date "2023-01-01" @default.
- W4312823189 modified "2023-10-11" @default.
- W4312823189 title "Caption-Aided Product Detection via Collaborative Pseudo-Label Harmonization" @default.
- W4312823189 cites W1536680647 @default.
- W4312823189 cites W1994488211 @default.
- W4312823189 cites W2016016818 @default.
- W4312823189 cites W2088049833 @default.
- W4312823189 cites W2124351162 @default.
- W4312823189 cites W2194775991 @default.
- W4312823189 cites W2604260814 @default.
- W4312823189 cites W2606831796 @default.
- W4312823189 cites W2610510511 @default.
- W4312823189 cites W2798748179 @default.
- W4312823189 cites W2813911573 @default.
- W4312823189 cites W2963150697 @default.
- W4312823189 cites W2963271314 @default.
- W4312823189 cites W2963516811 @default.
- W4312823189 cites W2963603913 @default.
- W4312823189 cites W2981613027 @default.
- W4312823189 cites W2982103617 @default.
- W4312823189 cites W2982905682 @default.
- W4312823189 cites W2989514282 @default.
- W4312823189 cites W2991023920 @default.
- W4312823189 cites W2991662170 @default.
- W4312823189 cites W2997544875 @default.
- W4312823189 cites W3023138874 @default.
- W4312823189 cites W3035725370 @default.
- W4312823189 cites W3104770333 @default.
- W4312823189 cites W3113075965 @default.
- W4312823189 cites W4290065491 @default.
- W4312823189 cites W4297971002 @default.
- W4312823189 cites W639708223 @default.
- W4312823189 doi "https://doi.org/10.1109/tmm.2022.3222653" @default.
- W4312823189 hasPublicationYear "2023" @default.
- W4312823189 type Work @default.
- W4312823189 citedByCount "0" @default.
- W4312823189 crossrefType "journal-article" @default.
- W4312823189 hasAuthorship W4312823189A5005325306 @default.
- W4312823189 hasAuthorship W4312823189A5037966954 @default.
- W4312823189 hasAuthorship W4312823189A5043757808 @default.
- W4312823189 hasAuthorship W4312823189A5047878798 @default.
- W4312823189 hasAuthorship W4312823189A5048961496 @default.
- W4312823189 hasAuthorship W4312823189A5049388503 @default.
- W4312823189 hasAuthorship W4312823189A5087043856 @default.
- W4312823189 hasConcept C119857082 @default.
- W4312823189 hasConcept C13280743 @default.
- W4312823189 hasConcept C154945302 @default.
- W4312823189 hasConcept C185798385 @default.
- W4312823189 hasConcept C204321447 @default.
- W4312823189 hasConcept C205649164 @default.
- W4312823189 hasConcept C23123220 @default.
- W4312823189 hasConcept C2524010 @default.
- W4312823189 hasConcept C33923547 @default.
- W4312823189 hasConcept C41008148 @default.
- W4312823189 hasConcept C90673727 @default.
- W4312823189 hasConceptScore W4312823189C119857082 @default.
- W4312823189 hasConceptScore W4312823189C13280743 @default.
- W4312823189 hasConceptScore W4312823189C154945302 @default.
- W4312823189 hasConceptScore W4312823189C185798385 @default.
- W4312823189 hasConceptScore W4312823189C204321447 @default.
- W4312823189 hasConceptScore W4312823189C205649164 @default.
- W4312823189 hasConceptScore W4312823189C23123220 @default.
- W4312823189 hasConceptScore W4312823189C2524010 @default.
- W4312823189 hasConceptScore W4312823189C33923547 @default.
- W4312823189 hasConceptScore W4312823189C41008148 @default.
- W4312823189 hasConceptScore W4312823189C90673727 @default.
- W4312823189 hasFunder F4320321001 @default.
- W4312823189 hasFunder F4320329791 @default.
- W4312823189 hasLocation W43128231891 @default.
- W4312823189 hasOpenAccess W4312823189 @default.
- W4312823189 hasPrimaryLocation W43128231891 @default.
- W4312823189 hasRelatedWork W112744582 @default.
- W4312823189 hasRelatedWork W1485630101 @default.
- W4312823189 hasRelatedWork W2070338563 @default.
- W4312823189 hasRelatedWork W2366644548 @default.
- W4312823189 hasRelatedWork W2376314740 @default.
- W4312823189 hasRelatedWork W2384888906 @default.
- W4312823189 hasRelatedWork W2498017833 @default.
- W4312823189 hasRelatedWork W2961085424 @default.
- W4312823189 hasRelatedWork W4306674287 @default.
- W4312823189 hasRelatedWork W4224009465 @default.
- W4312823189 hasVolume "25" @default.
- W4312823189 isParatext "false" @default.
- W4312823189 isRetracted "false" @default.
- W4312823189 workType "article" @default.