Retrieval-Guided and Semantically Grounded Image Captioning for Open-Domain Scenes

Recent image captioning methods based on pre-trained vision–language models can generate fluent and coherent descriptions, yet they still struggle in open-domain scenes that contain long-tail concepts, uncommon object combinations, and ambiguous visual evidence. Two limitations are especially import...

Descrizione completa

Dettagli Bibliografici

Autori principali:	Shanshan Lin, Xiaoxuan Xie, Zexian Yang, Chao Chen
Natura:	Artigo
Lingua:	Inglês
Pubblicazione:	MDPI AG 2026-05-01
Serie:	Mathematics
Soggetti:	image captioning open-domain captioning retrieval augmentation visual prompting semantic grounding vision–language models
Accesso online:	https://www.mdpi.com/2227-7390/14/10/1667
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!