LLM-enhanced Composed Image Retrieval: An Intent Uncertainty-aware Linguistic-Visual Dual Channel Matching Model

Composed image retrieval (CoIR) involves a multi-modal query of the reference image and modification text describing the desired changes, allowing users to express image retrieval intents flexibly and effectively. The key of CoIR lies in how to properly reason the search intent from the multi-modal...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	ACM transactions on information systems 2025-03, Vol.43 (2), p.1-30
Hauptverfasser:	Ge, Hongfei, Jiang, Yuanchun, Sun, Jianshan, Yuan, Kun, Liu, Yezheng
Format:	Artikel
Sprache:	eng
Schlagworte:	Image search Information systems
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!