arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.01082cs.CVcs.IRcs.LG
收藏

证据 Transformer 用于改进图像检索

Evidential Transformers for Improved Image Retrieval

ETH Zürich(苏黎世联邦理工学院);Texas A&M University(德克萨斯大学)

Danilo Dordevic, Suryansh Kumar

AI总结:

本文提出证据 Transformer,将概率方法融入图像检索,利用 GC ViT 架构,在 SOP 和 CUB-200-2011 数据集上取得新的最先进结果。

摘要:

我们引入了证据 Transformer(Evidential Transformer),一种基于不确定性的 Transformer 模型,用于改进和稳健的图像检索。在本文中,我们对基于内容的图像检索(CBIR)做出了多项贡献。我们将概率方法融入图像检索,获得了稳健且可靠的结果,其中证据分类(evidential classification)超越了基于多类分类的传统训练,作为深度度量学习的基线。此外,我们利用全局上下文视觉 Transformer(GC ViT)架构,在多个数据集上提升了最先进的检索结果。我们的实验结果一致地证明了我们方法的可靠性,在斯坦福在线产品(SOP)和 CUB-200-2011 数据集的所有测试设置中,为 CBIR 树立了新的基准。

英文摘要:

We introduce the Evidential Transformer, an uncertainty-driven transformer model for improved and robust image retrieval. In this paper, we make several contributions to content-based image retrieval (CBIR). We incorporate probabilistic methods into image retrieval, achieving robust and reliable results, with evidential classification surpassing traditional training based on multiclass classification as a baseline for deep metric learning. Furthermore, we improve the state-of-the-art retrieval results on several datasets by leveraging the Global Context Vision Transformer (GC ViT) architecture. Our experimental results consistently demonstrate the reliability of our approach, setting a new benchmark in CBIR in all test settings on the Stanford Online Products (SOP) and CUB-200-2011 datasets.

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。

↑