arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2503.11481cs.CV
收藏

T2I-FineEval:用于文本到图像评估的细粒度组合性指标

T2I-FineEval: Fine-Grained Compositional Metric for Text-to-Image Evaluation

Sharif University of Technology(谢里夫理工大学)

Seyed Mohammad Hadi Hosseini, Amir Mohammad Izadi, Ali Abdollahi, Armin Saghafian, Mahdieh Soleymani Baghshah

AI总结:

针对文本到图像模型在属性绑定和空间关系等组合性上的评估不足,本文提出 T2I-FineEval,通过分解图像组件与细粒度 VQA 问题实现更准确评估,性能超过现有最先进指标。

摘要:

尽管近期的文本到图像生成模型已取得令人瞩目的性能,但它们在捕捉提示词的组合复杂性方面仍常显吃力,这些复杂性包括属性绑定以及不同实体之间的空间关系。这种错位并未被 CLIPScore 等常用评估指标揭示出来。近期的工作提出了利用视觉问答(VQA)的评估指标,其方式是将提示词分解为关于所生成图像的问题,以进行更稳健的组合性评估。尽管这些方法与人类评估更加一致,但它们仍未能完全覆盖图像内部的组合性。为解决这一问题,我们提出了一种新型指标,该指标将图像分解为若干组成部分,并将文本分解为关于所生成图像的细粒度问题以进行评估。我们的方法优于此前最先进的指标,证明了其在评估文本到图像生成模型方面的有效性。代码可在 https://github.com/hadi-hosseini/T2I-FineEval 获取。

英文摘要:

Although recent text-to-image generative models have achieved impressive performance, they still often struggle with capturing the compositional complexities of prompts including attribute binding, and spatial relationships between different entities. This misalignment is not revealed by common evaluation metrics such as CLIPScore. Recent works have proposed evaluation metrics that utilize Visual Question Answering (VQA) by decomposing prompts into questions about the generated image for more robust compositional evaluation. Although these methods align better with human evaluations, they still fail to fully cover the compositionality within the image. To address this, we propose a novel metric that breaks down images into components, and texts into fine-grained questions about the generated image for evaluation. Our method outperforms previous state-of-the-art metrics, demonstrating its effectiveness in evaluating text-to-image generative models. Code is available at https://github.com/hadi-hosseini/ T2I-FineEval.

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。

↑