Reproducing "Show, attend and tell: Neural image caption generation with visual attention"

Liu, Haixia; Brailsford, Tim

Abstract

This paper replicates the experiment presented in the work of Xu et al. [1], and
examines errors in the generated captions. The analysis of the identified errors aims to provide
deeper insight into the underlying causes. This study also encompasses subsequent experiments
aiming at investigating the feasibility of rectifying these errors via a post-processing stage. Image
recognition and object detection models, as well as a language probability computational model
were explored. The findings presented in this paper aim to contribute towards the overarching
objective of Explainable Artificial Intelligence (XAI), thereby providing potential pathways to
improve image captioning.

• Presentation Conference Type: Conference Paper (published)
• Conference Name: The 16th International Conference on Computer and Electrical Engineering
• Acceptance Date: Apr 30, 2023
• Online Publication Date: Sep 13, 2023
• Publication Date: Sep 13, 2023
• Deposit Date: May 17, 2023
• Publicly Available Date: Oct 3, 2023
• Journal: Journal of Physics: Conference Series
• Print ISSN: 1742-6588
• Electronic ISSN: 1742-6596
• Publisher: IOP Publishing
• Peer Reviewed: Peer Reviewed
• Volume: 2589
• Issue: 1
• Article Number: 012012
• Series ISSN: 1742-6596
• DOI: https://doi.org/10.1088/1742-6596/2589/1/012012
• Public URL: https://uwe-repository.worktribe.com/output/10797312