arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2504.02496cs.CVcs.MM
收藏

基于记忆差异编码与注意力的分组区分性图像描述生成

Group-based Distinctive Image Captioning with Memory Difference Encoding and Attention

City University of Hong Kong(香港城市大学);Alibaba Group(阿里巴巴集团);University of Chinese Academy of Sciences(中国科学院大学)

Jiuniu Wang, Wenjia Xu, Qingzhong Wang, Antoni B. Chan

AI总结:

本文提出基于分组的差异区分性图像描述方法,通过GDMA模块比较相似图像组内对象特征并优先关注独特特征,结合区分性词汇指导解码,显著提升描述区分性且不过度牺牲准确性。

摘要:

图像描述生成领域的最新进展主要通过大幅增加数据集和模型规模来提升准确性。尽管传统描述模型在BLEU、CIDEr和SPICE等既有指标上表现优异,但描述区分目标图像与其他相似图像的能力仍被探索不足。为生成区分性描述,少数先驱采用对比学习或对真实描述进行重加权。然而,这些方法常忽略相似图像组(如同一相册中的物品、属性或细粒度事件)内对象间的关系。本文引入一种增强图像描述区分性的新方法,即基于分组的差异区分性描述方法,该方法在视觉上将每张图像与相似组内其他图像进行比较,凸显各图像的独特性。具体而言,我们引入基于分组的差异记忆注意力模块,旨在识别并强调图像中在其图像组内具有独特区分性的对象特征,即那些与其他图像中对象相似度低的特征。此机制确保在为图像生成描述时优先处理这些独特对象特征,从而增强生成描述的区分性。为进一步优化此过程,我们从真实描述中选取区分性词汇以指导语言解码器和GDMA模块。此外,我们提出新评估指标区分性词率,以定量评估描述区分性。定量结果表明,所提方法显著提升多个基线模型的区分性,并在区分性上达到SOTA性能且未过度牺牲准确性。

英文摘要:

Recent advances in image captioning have focused on enhancing accuracy by substantially increasing the dataset and model size. While conventional captioning models exhibit high performance on established metrics such as BLEU, CIDEr, and SPICE, the capability of captions to distinguish the target image from other similar images is under-explored. To generate distinctive captions, a few pioneers employed contrastive learning or re-weighted the ground-truth captions. However, these approaches often overlook the relationships among objects in a similar image group (e.g., items or properties within the same album or fine-grained events). In this paper, we introduce a novel approach to enhance the distinctiveness of image captions, namely Group-based Differential Distinctive Captioning Method, which visually compares each image with other images in one similar group and highlights the uniqueness of each image. In particular, we introduce a Group-based Differential Memory Attention (GDMA) module, designed to identify and emphasize object features in an image that are uniquely distinguishable within its image group, i.e., those exhibiting low similarity with objects in other images. This mechanism ensures that such unique object features are prioritized during caption generation for the image, thereby enhancing the distinctiveness of the resulting captions. To further refine this process, we select distinctive words from the ground-truth captions to guide both the language decoder and the GDMA module. Additionally, we propose a new evaluation metric, the Distinctive Word Rate (DisWordRate), to quantitatively assess caption distinctiveness. Quantitative results indicate that the proposed method significantly improves the distinctiveness of several baseline models, and achieves state-of-the-art performance on distinctiveness while not excessively sacrificing accuracy...

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑