FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952418266> ?p ?o ?g. }

• W2952418266 abstract "Due to the fact that it is prohibitively expensive to completely annotate visual relationships, i.e., the (obj1, rel, obj2) triplets, relationship models are inevitably biased to object classes of limited pairwise patterns, leading to poor generalization to rare or unseen object combinations. Therefore, we are interested in learning object-agnostic visual features for more generalizable relationship models. By agnostic, we mean that the feature is less likely biased to the classes of paired objects. To alleviate the bias, we propose a novel texttt{Shuffle-Then-Assemble} pre-training strategy. First, we discard all the triplet relationship annotations in an image, leaving two unpaired object domains without obj1-obj2 alignment. Then, our feature learning is to recover possible obj1-obj2 pairs. In particular, we design a cycle of residual transformations between the two domains, to capture shared but not object-specific visual patterns. Extensive experiments on two visual relationship benchmarks show that by using our pre-trained features, naive relationship models can be consistently improved and even outperform other state-of-the-art relationship models. Code has been made available at: url{https://github.com/yangxuntu/vrd}." @default.
• W2952418266 created "2019-06-27" @default.
• W2952418266 creator A5017901486 @default.
• W2952418266 creator A5042324027 @default.
• W2952418266 creator A5064752243 @default.
• W2952418266 date "2018-08-01" @default.
• W2952418266 modified "2023-10-17" @default.
• W2952418266 title "Shuffle-Then-Assemble: Learning Object-Agnostic Visual Relationship Features" @default.
• W2952418266 cites W1517303303 @default.
• W2952418266 cites W1522301498 @default.
• W2952418266 cites W1584193343 @default.
• W2952418266 cites W1861492603 @default.
• W2952418266 cites W1892016050 @default.
• W2952418266 cites W1921523184 @default.
• W2952418266 cites W2046589395 @default.
• W2952418266 cites W2049705550 @default.
• W2952418266 cites W2077069816 @default.
• W2952418266 cites W2108598243 @default.
• W2952418266 cites W2149933564 @default.
• W2952418266 cites W2163922914 @default.
• W2952418266 cites W2169393274 @default.
• W2952418266 cites W2190656909 @default.
• W2952418266 cites W2194775991 @default.
• W2952418266 cites W2214124602 @default.
• W2952418266 cites W2274287116 @default.
• W2952418266 cites W2277195237 @default.
• W2952418266 cites W2321533354 @default.
• W2952418266 cites W2412320034 @default.
• W2952418266 cites W2423576022 @default.
• W2952418266 cites W2463955103 @default.
• W2952418266 cites W2513221924 @default.
• W2952418266 cites W2550553598 @default.
• W2952418266 cites W2558535589 @default.
• W2952418266 cites W2558661413 @default.
• W2952418266 cites W2591644541 @default.
• W2952418266 cites W2593414223 @default.
• W2952418266 cites W2605736949 @default.
• W2952418266 cites W2608015370 @default.
• W2952418266 cites W2740118378 @default.
• W2952418266 cites W2754191212 @default.
• W2952418266 cites W2771002095 @default.
• W2952418266 cites W2771558241 @default.
• W2952418266 cites W2777602943 @default.
• W2952418266 cites W2790777763 @default.
• W2952418266 cites W2798376494 @default.
• W2952418266 cites W2883170015 @default.
• W2952418266 cites W2887585070 @default.
• W2952418266 cites W2949831469 @default.
• W2952418266 cites W2951165067 @default.
• W2952418266 cites W2951343884 @default.
• W2952418266 cites W2951939904 @default.
• W2952418266 cites W2952020101 @default.
• W2952418266 cites W2953106684 @default.
• W2952418266 cites W2953238423 @default.
• W2952418266 cites W2962764817 @default.
• W2952418266 cites W2962793481 @default.
• W2952418266 cites W2962808524 @default.
• W2952418266 cites W2962835968 @default.
• W2952418266 cites W2962949934 @default.
• W2952418266 cites W2963107255 @default.
• W2952418266 cites W2963420272 @default.
• W2952418266 cites W2963536419 @default.
• W2952418266 cites W2963622213 @default.
• W2952418266 cites W2963649796 @default.
• W2952418266 cites W2963684088 @default.
• W2952418266 cites W2963955422 @default.
• W2952418266 cites W343636949 @default.
• W2952418266 doi "https://doi.org/10.48550/arxiv.1808.00171" @default.
• W2952418266 hasPublicationYear "2018" @default.
• W2952418266 type Work @default.
• W2952418266 sameAs 2952418266 @default.
• W2952418266 citedByCount "0" @default.
• W2952418266 crossrefType "posted-content" @default.
• W2952418266 hasAuthorship W2952418266A5017901486 @default.
• W2952418266 hasAuthorship W2952418266A5042324027 @default.
• W2952418266 hasAuthorship W2952418266A5064752243 @default.
• W2952418266 hasBestOaLocation W29524182661 @default.
• W2952418266 hasConcept C115961682 @default.
• W2952418266 hasConcept C119857082 @default.
• W2952418266 hasConcept C134306372 @default.
• W2952418266 hasConcept C138885662 @default.
• W2952418266 hasConcept C153180895 @default.
• W2952418266 hasConcept C154945302 @default.
• W2952418266 hasConcept C177148314 @default.
• W2952418266 hasConcept C177264268 @default.
• W2952418266 hasConcept C184898388 @default.
• W2952418266 hasConcept C199360897 @default.
• W2952418266 hasConcept C2776401178 @default.
• W2952418266 hasConcept C2776760102 @default.
• W2952418266 hasConcept C2779542340 @default.
• W2952418266 hasConcept C2781238097 @default.
• W2952418266 hasConcept C33923547 @default.
• W2952418266 hasConcept C41008148 @default.
• W2952418266 hasConcept C41895202 @default.
• W2952418266 hasConcept C80444323 @default.
• W2952418266 hasConceptScore W2952418266C115961682 @default.
• W2952418266 hasConceptScore W2952418266C119857082 @default.
• W2952418266 hasConceptScore W2952418266C134306372 @default.
• W2952418266 hasConceptScore W2952418266C138885662 @default.
• W2952418266 hasConceptScore W2952418266C153180895 @default.

• next