Eye-movement-prompted large image captioning model

Open Webpage

Zheng Yang, Bing Han, Xinbo Gao, Zhi-Hui Zhan

Published: 01 Jan 2025, Last Modified: 13 May 2025Pattern Recognit. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•A dataset was built to explore relationships between human vision and language.•We propose an eye-movement-prompted large image captioning model in this paper.•A GNN-based module is designed to explore useful features of eye-movement data.•The effectiveness and interpretability of EMLIC were verified on two datasets.