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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

This work focuses on sign language retrieval-a recently proposed task forsign language understanding. Sign language retrieval consists of two sub-tasks:text-to-sign-video (T2V) retrieval and sign-video-to-text (V2T) retrieval.Different from traditional video-text retrieval, sign language videos, not onlycontain visual signals but also carry abundant semantic meanings by themselvesdue to the fact that sign languages are also natural languages. Consideringthis character, we formulate sign language retrieval as a cross-lingualretrieval problem as well as a video-text retrieval task. Concretely, we takeinto account the linguistic properties of both sign languages and naturallanguages, and simultaneously identify the fine-grained cross-lingual (i.e.,sign-to-word) mappings while contrasting the texts and the sign videos in ajoint embedding space. This process is termed as cross-lingual contrastivelearning. Another challenge is raised by the data scarcity issue-sign languagedatasets are orders of magnitude smaller in scale than that of speechrecognition. We alleviate this issue by adopting a domain-agnostic sign encoderpre-trained on large-scale sign videos into the target domain viapseudo-labeling. Our framework, termed as domain-aware sign language retrievalvia Cross-lingual Contrastive learning or CiCo for short, outperforms thepioneering method by large margins on various datasets, e.g., +22.4 T2V and+28.0 V2T R@1 improvements on How2Sign dataset, and +13.7 T2V and +17.1 V2T R@1improvements on PHOENIX-2014T dataset. Code and models are available at:https://github.com/FangyunWei/SLRT.

CiCo: Domain-Aware Sign Language Retrieval via Cross-Lingual Contrastive Learning