Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204380095> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W3204380095 endingPage "108358" @default.
- W3204380095 startingPage "108358" @default.
- W3204380095 abstract "Image captioning is a hot research topic bridging computer vision and natural language processing during the past several decades. It has achieved great progress with the help of large-scale datasets and deep learning techniques. Though the variety of image captioning models (ICMs), the performance of ICMs have got stuck in a bottleneck judging from the publicly published results. Considering the marginal performance gains brought by recent ICMs, we raise the following question: “what about the performances of the recent ICMs achieve on in-the-wild images? To clarify this question, we compare existing ICMs by evaluating their generalization ability. Specifically, we propose a novel method based on maximum discrepancy competition to diagnose existing ICMs. Firstly, we establish a new test set containing only informative images selected by adopting maximum discrepancy competition on the existing ICMs, from an arbitrary large-scale raw image set. Secondly, a small-scale and low-cost subjective annotation experiment is conducted on the new test set. Thirdly, we rank the generalization ability of the existing ICMs by comparing their performances on the new test set. Finally, the keys of different ICMs are demonstrated based on a detailed analysis of experimental results. Our analysis yields several interesting findings, including that 1) Using simultaneously low- and high-level object features may be an effective tool to boost the generalization ability for the Transformer based ICMs. 2) Self-attention mechanism may provide better modelling ability for inter- and intra-modal data than other attention-based mechanisms. 3) Constructing an ICM with a multistage language decoder may be a promising way to improve its performance." @default.
- W3204380095 created "2021-10-11" @default.
- W3204380095 creator A5023363049 @default.
- W3204380095 creator A5027250401 @default.
- W3204380095 creator A5030028635 @default.
- W3204380095 creator A5030344445 @default.
- W3204380095 creator A5059718700 @default.
- W3204380095 creator A5063013411 @default.
- W3204380095 date "2022-02-01" @default.
- W3204380095 modified "2023-09-28" @default.
- W3204380095 title "Revisiting image captioning via maximum discrepancy competition" @default.
- W3204380095 cites W1969616664 @default.
- W3204380095 cites W2048366140 @default.
- W3204380095 cites W2117539524 @default.
- W3204380095 cites W2126226185 @default.
- W3204380095 cites W2128272608 @default.
- W3204380095 cites W2185175083 @default.
- W3204380095 cites W2277195237 @default.
- W3204380095 cites W2296385829 @default.
- W3204380095 cites W2463955103 @default.
- W3204380095 cites W2906729185 @default.
- W3204380095 cites W2914163459 @default.
- W3204380095 cites W2966314676 @default.
- W3204380095 cites W2979739834 @default.
- W3204380095 cites W3007051430 @default.
- W3204380095 doi "https://doi.org/10.1016/j.patcog.2021.108358" @default.
- W3204380095 hasPublicationYear "2022" @default.
- W3204380095 type Work @default.
- W3204380095 sameAs 3204380095 @default.
- W3204380095 citedByCount "8" @default.
- W3204380095 countsByYear W32043800952022 @default.
- W3204380095 countsByYear W32043800952023 @default.
- W3204380095 crossrefType "journal-article" @default.
- W3204380095 hasAuthorship W3204380095A5023363049 @default.
- W3204380095 hasAuthorship W3204380095A5027250401 @default.
- W3204380095 hasAuthorship W3204380095A5030028635 @default.
- W3204380095 hasAuthorship W3204380095A5030344445 @default.
- W3204380095 hasAuthorship W3204380095A5059718700 @default.
- W3204380095 hasAuthorship W3204380095A5063013411 @default.
- W3204380095 hasConcept C115961682 @default.
- W3204380095 hasConcept C119857082 @default.
- W3204380095 hasConcept C134306372 @default.
- W3204380095 hasConcept C149635348 @default.
- W3204380095 hasConcept C153180895 @default.
- W3204380095 hasConcept C154945302 @default.
- W3204380095 hasConcept C157657479 @default.
- W3204380095 hasConcept C177148314 @default.
- W3204380095 hasConcept C177264268 @default.
- W3204380095 hasConcept C199360897 @default.
- W3204380095 hasConcept C2780513914 @default.
- W3204380095 hasConcept C33923547 @default.
- W3204380095 hasConcept C41008148 @default.
- W3204380095 hasConceptScore W3204380095C115961682 @default.
- W3204380095 hasConceptScore W3204380095C119857082 @default.
- W3204380095 hasConceptScore W3204380095C134306372 @default.
- W3204380095 hasConceptScore W3204380095C149635348 @default.
- W3204380095 hasConceptScore W3204380095C153180895 @default.
- W3204380095 hasConceptScore W3204380095C154945302 @default.
- W3204380095 hasConceptScore W3204380095C157657479 @default.
- W3204380095 hasConceptScore W3204380095C177148314 @default.
- W3204380095 hasConceptScore W3204380095C177264268 @default.
- W3204380095 hasConceptScore W3204380095C199360897 @default.
- W3204380095 hasConceptScore W3204380095C2780513914 @default.
- W3204380095 hasConceptScore W3204380095C33923547 @default.
- W3204380095 hasConceptScore W3204380095C41008148 @default.
- W3204380095 hasLocation W32043800951 @default.
- W3204380095 hasOpenAccess W3204380095 @default.
- W3204380095 hasPrimaryLocation W32043800951 @default.
- W3204380095 hasRelatedWork W2795359650 @default.
- W3204380095 hasRelatedWork W2803367139 @default.
- W3204380095 hasRelatedWork W2909133588 @default.
- W3204380095 hasRelatedWork W2923366293 @default.
- W3204380095 hasRelatedWork W2989932438 @default.
- W3204380095 hasRelatedWork W3008515501 @default.
- W3204380095 hasRelatedWork W3041777105 @default.
- W3204380095 hasRelatedWork W3102877762 @default.
- W3204380095 hasRelatedWork W3201070945 @default.
- W3204380095 hasRelatedWork W4284710034 @default.
- W3204380095 hasVolume "122" @default.
- W3204380095 isParatext "false" @default.
- W3204380095 isRetracted "false" @default.
- W3204380095 magId "3204380095" @default.
- W3204380095 workType "article" @default.