Word to Sentence Visual Semantic Similarity for Caption Generation: Lessons Learned

TL;DR

This paper introduces a visual semantic similarity approach to improve image captioning by selecting captions most related to the image, rather than the most probable output, enhancing caption relevance.

Contribution

It presents a post-processing method that uses visual semantic measures to select more accurate captions for images, applicable to any captioning system.

Findings

Improved caption relevance through semantic matching

Applicable as a post-processing step for existing systems

Enhances image-caption alignment accuracy

Abstract

This paper focuses on enhancing the captions generated by image-caption generation systems. We propose an approach for improving caption generation systems by choosing the most closely related output to the image rather than the most likely output produced by the model. Our model revises the language generation output beam search from a visual context perspective. We employ a visual semantic measure in a word and sentence level manner to match the proper caption to the related information in the image. The proposed approach can be applied to any caption system as a post-processing based method.