MeaCap: Memory-Augmented Zero-shot Image Captioning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zeng, Zequn, Xie, Yan, Zhang, Hao, Chen, Chiyu, Wang, Zhengjue, Chen, Bo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917605992824832
author Zeng, Zequn
Xie, Yan
Zhang, Hao
Chen, Chiyu
Wang, Zhengjue
Chen, Bo
author_facet Zeng, Zequn
Xie, Yan
Zhang, Hao
Chen, Chiyu
Wang, Zhengjue
Chen, Bo
contents Zero-shot image captioning (IC) without well-paired image-text data can be divided into two categories, training-free and text-only-training. Generally, these two types of methods realize zero-shot IC by integrating pretrained vision-language models like CLIP for image-text similarity evaluation and a pre-trained language model (LM) for caption generation. The main difference between them is whether using a textual corpus to train the LM. Though achieving attractive performance w.r.t. some metrics, existing methods often exhibit some common drawbacks. Training-free methods tend to produce hallucinations, while text-only-training often lose generalization capability. To move forward, in this paper, we propose a novel Memory-Augmented zero-shot image Captioning framework (MeaCap). Specifically, equipped with a textual memory, we introduce a retrieve-then-filter module to get key concepts that are highly related to the image. By deploying our proposed memory-augmented visual-related fusion score in a keywords-to-sentence LM, MeaCap can generate concept-centered captions that keep high consistency with the image with fewer hallucinations and more world-knowledge. The framework of MeaCap achieves the state-of-the-art performance on a series of zero-shot IC settings. Our code is available at https://github.com/joeyz0z/MeaCap.
format Preprint
id arxiv_https___arxiv_org_abs_2403_03715
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MeaCap: Memory-Augmented Zero-shot Image Captioning
Zeng, Zequn
Xie, Yan
Zhang, Hao
Chen, Chiyu
Wang, Zhengjue
Chen, Bo
Computer Vision and Pattern Recognition
Zero-shot image captioning (IC) without well-paired image-text data can be divided into two categories, training-free and text-only-training. Generally, these two types of methods realize zero-shot IC by integrating pretrained vision-language models like CLIP for image-text similarity evaluation and a pre-trained language model (LM) for caption generation. The main difference between them is whether using a textual corpus to train the LM. Though achieving attractive performance w.r.t. some metrics, existing methods often exhibit some common drawbacks. Training-free methods tend to produce hallucinations, while text-only-training often lose generalization capability. To move forward, in this paper, we propose a novel Memory-Augmented zero-shot image Captioning framework (MeaCap). Specifically, equipped with a textual memory, we introduce a retrieve-then-filter module to get key concepts that are highly related to the image. By deploying our proposed memory-augmented visual-related fusion score in a keywords-to-sentence LM, MeaCap can generate concept-centered captions that keep high consistency with the image with fewer hallucinations and more world-knowledge. The framework of MeaCap achieves the state-of-the-art performance on a series of zero-shot IC settings. Our code is available at https://github.com/joeyz0z/MeaCap.
title MeaCap: Memory-Augmented Zero-shot Image Captioning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.03715