Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378505318> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4378505318 abstract "A great deal of progress has been made in image captioning, driven by research into how to encode the image using pre-trained models. This includes visual encodings (e.g. image grid features or detected objects) and more recently textual encodings (e.g. image tags or text descriptions of image regions). As more advanced encodings are available and incorporated, it is natural to ask: how to efficiently and effectively leverage the heterogeneous set of encodings? In this paper, we propose to regard the encodings as augmented views of the input image. The image captioning model encodes each view independently with a shared encoder efficiently, and a contrastive loss is incorporated across the encoded views in a novel way to improve their representation quality and the model's data efficiency. Our proposed hierarchical decoder then adaptively weighs the encoded views according to their effectiveness for caption generation by first aggregating within each view at the token level, and then across views at the view level. We demonstrate significant performance improvements of +5.6% CIDEr on MS-COCO and +12.9% CIDEr on Flickr30k compared to state of the arts, and conduct rigorous analyses to demonstrate the importance of each part of our design." @default.
- W4378505318 created "2023-05-27" @default.
- W4378505318 creator A5018408927 @default.
- W4378505318 creator A5088892134 @default.
- W4378505318 date "2023-05-25" @default.
- W4378505318 modified "2023-09-30" @default.
- W4378505318 title "HAAV: Hierarchical Aggregation of Augmented Views for Image Captioning" @default.
- W4378505318 doi "https://doi.org/10.48550/arxiv.2305.16295" @default.
- W4378505318 hasPublicationYear "2023" @default.
- W4378505318 type Work @default.
- W4378505318 citedByCount "0" @default.
- W4378505318 crossrefType "posted-content" @default.
- W4378505318 hasAuthorship W4378505318A5018408927 @default.
- W4378505318 hasAuthorship W4378505318A5088892134 @default.
- W4378505318 hasBestOaLocation W43785053181 @default.
- W4378505318 hasConcept C104317684 @default.
- W4378505318 hasConcept C111919701 @default.
- W4378505318 hasConcept C115961682 @default.
- W4378505318 hasConcept C118505674 @default.
- W4378505318 hasConcept C153083717 @default.
- W4378505318 hasConcept C154945302 @default.
- W4378505318 hasConcept C157657479 @default.
- W4378505318 hasConcept C177264268 @default.
- W4378505318 hasConcept C17744445 @default.
- W4378505318 hasConcept C185592680 @default.
- W4378505318 hasConcept C199360897 @default.
- W4378505318 hasConcept C199539241 @default.
- W4378505318 hasConcept C2776359362 @default.
- W4378505318 hasConcept C31972630 @default.
- W4378505318 hasConcept C38652104 @default.
- W4378505318 hasConcept C41008148 @default.
- W4378505318 hasConcept C48145219 @default.
- W4378505318 hasConcept C55493867 @default.
- W4378505318 hasConcept C66746571 @default.
- W4378505318 hasConcept C94625758 @default.
- W4378505318 hasConceptScore W4378505318C104317684 @default.
- W4378505318 hasConceptScore W4378505318C111919701 @default.
- W4378505318 hasConceptScore W4378505318C115961682 @default.
- W4378505318 hasConceptScore W4378505318C118505674 @default.
- W4378505318 hasConceptScore W4378505318C153083717 @default.
- W4378505318 hasConceptScore W4378505318C154945302 @default.
- W4378505318 hasConceptScore W4378505318C157657479 @default.
- W4378505318 hasConceptScore W4378505318C177264268 @default.
- W4378505318 hasConceptScore W4378505318C17744445 @default.
- W4378505318 hasConceptScore W4378505318C185592680 @default.
- W4378505318 hasConceptScore W4378505318C199360897 @default.
- W4378505318 hasConceptScore W4378505318C199539241 @default.
- W4378505318 hasConceptScore W4378505318C2776359362 @default.
- W4378505318 hasConceptScore W4378505318C31972630 @default.
- W4378505318 hasConceptScore W4378505318C38652104 @default.
- W4378505318 hasConceptScore W4378505318C41008148 @default.
- W4378505318 hasConceptScore W4378505318C48145219 @default.
- W4378505318 hasConceptScore W4378505318C55493867 @default.
- W4378505318 hasConceptScore W4378505318C66746571 @default.
- W4378505318 hasConceptScore W4378505318C94625758 @default.
- W4378505318 hasLocation W43785053181 @default.
- W4378505318 hasOpenAccess W4378505318 @default.
- W4378505318 hasPrimaryLocation W43785053181 @default.
- W4378505318 hasRelatedWork W2005185696 @default.
- W4378505318 hasRelatedWork W2359640100 @default.
- W4378505318 hasRelatedWork W2407414618 @default.
- W4378505318 hasRelatedWork W2547835662 @default.
- W4378505318 hasRelatedWork W2795359650 @default.
- W4378505318 hasRelatedWork W2923366293 @default.
- W4378505318 hasRelatedWork W3008321871 @default.
- W4378505318 hasRelatedWork W3008515501 @default.
- W4378505318 hasRelatedWork W3163104128 @default.
- W4378505318 hasRelatedWork W4284710034 @default.
- W4378505318 isParatext "false" @default.
- W4378505318 isRetracted "false" @default.
- W4378505318 workType "article" @default.