Text-based image retrieval system using semantic visual content for re-ranking

Berkay Topçu

doi:10.1016/j.engappai.2025.111770

Text-based image retrieval system using semantic visual content for re-ranking

2025-07-30

0

PRE

AI

DOI：10.1016/j.engappai.2025.111770

Original

Original request for help

Save

Abstract

En 中文

• Overview of the workings of a commercial image retrieval system, filling a gap in existing literature. • Text-based retrieval is significantly improved using content analysis via deep image embeddings. • Filtering irrelevant images is possible by clustering candidate images based on embeddings. • Re-ranking is possible based on the novel idea of a “master image” of a query.

Keywords:

image retrieval

deep image embeddings

clustering

re-ranking

master image

Text-based image retrieval system using semantic visual content for re-ranking

Abstract

Journal

Organization