English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Image-text matching is gaining a leading role among tasks involving the jointunderstanding of vision and language. In literature, this task is often used asa pre-training objective to forge architectures able to jointly deal withimages and texts. Nonetheless, it has a direct downstream application:cross-modal retrieval, which consists in finding images related to a givenquery text or vice-versa. Solving this task is of critical importance incross-modal search engines. Many recent methods proposed effective solutions tothe image-text matching problem, mostly using recent large vision-language (VL)Transformer networks. However, these models are often computationallyexpensive, especially at inference time. This prevents their adoption inlarge-scale cross-modal retrieval scenarios, where results should be providedto the user almost instantaneously. In this paper, we propose to fill in thegap between effectiveness and efficiency by proposing an ALign And DIstillNetwork (ALADIN). ALADIN first produces high-effective scores by aligning atfine-grained level images and texts. Then, it learns a shared embedding space -where an efficient kNN search can be performed - by distilling the relevancescores obtained from the fine-grained alignments. We obtained remarkableresults on MS-COCO, showing that our method can compete with state-of-the-artVL Transformers while being almost 90 times faster. The code for reproducingour results is available at https://github.com/mesnico/ALADIN.

ALADIN: Distilling Fine-grained Alignment Scores for Efficient Image-Text Matching and Retrieval