发表机构
Boston University; Creatify AI（波士顿大学; Creatify AI）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究推出图像生成与编辑相对物体尺度基准GenScale，设计序数评判器，提出无需修改源生成器的后处理智能体Rescale，实验显示现有顶尖模型难感知相对尺度，Rescale可提升尺度合理性。
AI 中文摘要
现代图像生成与编辑系统能够生成与提示词对齐的逼真图像，但仍常将熟悉物体渲染为不符合常理的相对尺寸。为衡量这一失效模式，我们推出GenScale，一个针对图像生成与编辑中真实世界相对物体尺度的基准测试及评估协议。GenScale包含900个图像级条目，以及1643组锚点-目标尺度关系，涵盖常见物体生成、带度量尺寸的人与产品生成，以及对失败生成结果的尺度修正。我们还设计了经人工校准的序数评判器，用于可扩展的成对尺度评估。最后，我们推出Rescale，一个与模型无关的后处理智能体，用于局部尺度修正而无需修改源生成器。实验表明，最先进的图像生成器与编辑器目前仍无法可靠感知相对尺度，而Rescale能持续提升生成与编辑图像的尺度合理性。综上，GenScale确立了相对物体尺度作为图像生成系统一项可区分、可测量且可落地的能力。
英文摘要
Modern image generation and editing systems can produce photorealistic, prompt-aligned images, but still often render familiar objects at implausible relative sizes. To measure this failure mode, we introduce GenScale, a benchmark and evaluation protocol for real-world relative object scale in image generation and editing. GenScale contains 900 image-level entries and 1,643 pairwise anchor-target scale relations across common-object generation, human-product generation with metric dimensions, and scale correction from failed generations. We further design a human-calibrated ordinal judge for scalable pairwise scale evaluation. Last but not the least, we introduce Rescale, a model-agnostic post-processing agent for localized scale correction without modifying the source generator. Experiments reveal that state-of-the-art image generators and editors cannot reliably observe relative scale yet, while Rescale consistently improves scale plausibility across generated and edited images. Together, GenScale establishes relative object scale as a distinct, measurable, and actionable capability for image generation systems.
