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

Visual Semantic Embedding (VSE) is a dominant approach for vision-languageretrieval, which aims at learning a deep embedding space such that visual dataare embedded close to their semantic text labels or descriptions. Recent VSEmodels use complex methods to better contextualize and aggregate multi-modalfeatures into holistic embeddings. However, we discover that surprisinglysimple (but carefully selected) global pooling functions (e.g., max pooling)outperform those complex models, across different feature extractors. Despiteits simplicity and effectiveness, seeking the best pooling function fordifferent data modality and feature extractor is costly and tedious, especiallywhen the size of features varies (e.g., text, video). Therefore, we propose aGeneralized Pooling Operator (GPO), which learns to automatically adapt itselfto the best pooling strategy for different features, requiring no manual tuningwhile staying effective and efficient. We extend the VSE model using thisproposed GPO and denote it as VSE$\infty$.  Without bells and whistles, VSE$\infty$ outperforms previous VSE methodssignificantly on image-text retrieval benchmarks across popular featureextractors. With a simple adaptation, variants of VSE$\infty$ furtherdemonstrate its strength by achieving the new state of the art on twovideo-text retrieval datasets. Comprehensive experiments and visualizationsconfirm that GPO always discovers the best pooling strategy and can be aplug-and-play feature aggregation module for standard VSE models. Code andpre-trained models are available at https://vse-infty.github.io.

Learning the Best Pooling Strategy for Visual Semantic Embedding