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

Quality feature representation is key to instance image retrieval. To attainit, existing methods usually resort to a deep model pre-trained on benchmarkdatasets or even fine-tune the model with a task-dependent labelled auxiliarydataset. Although achieving promising results, this approach is restricted bytwo issues: 1) the domain gap between benchmark datasets and the dataset of agiven retrieval task; 2) the required auxiliary dataset cannot be readilyobtained. In light of this situation, this work looks into a different approachwhich has not been well investigated for instance image retrieval previously:{can we learn feature representation \textit{specific to} a given retrievaltask in order to achieve excellent retrieval?} Our finding is encouraging. Byadding an object proposal generator to generate image regions forself-supervised learning, the investigated approach can successfully learnfeature representation specific to a given dataset for retrieval. Thisrepresentation can be made even more effective by boosting it with imagesimilarity information mined from the dataset. As experimentally validated,such a simple ``self-supervised learning + self-boosting'' approach can wellcompete with the relevant state-of-the-art retrieval methods. Ablation study isconducted to show the appealing properties of this approach and its limitationon generalisation across datasets.

Instance Image Retrieval by Learning Purely From Within the Dataset