arXiv 2510.176812026-01-04cs.CVcs.AI

PICABench: 我们距离物理真实图像编辑还有多远？

PICABench: How Far Are We from Physically Realistic Image Editing?

Shanghai Jiao Tong University（上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；CUHK MMLab（香港中文大学多媒体实验室）；Krea AI；Beihang University（北京航空航天大学）；Tongyi Lab（通义实验室）；USTC（中国科学技术大学）；The University of Hong Kong（香港大学）

Yuandong Pu, Le Zhuo, Songhao Han, Jinbo Xing, Kaiwen Zhu, Shuo Cao, Bin Fu, Si Liu, Hongsheng Li, Yu Qiao, Wenlong Zhang, Xi Chen, Yihao Liu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

PICABench通过系统评估物理真实感，探索图像编辑中物理效应的挑战与解决方案。

摘要：

图像编辑近年来取得了显著进展。现代编辑模型已经能够遵循复杂的指令来操控原始内容。然而，在完成编辑指令之外，伴随的物理效应是生成真实感的关键。例如，移除一个物体也应移除其阴影、反射以及与附近物体的相互作用。不幸的是，现有模型和基准主要关注指令完成，却忽略了这些物理效应。因此，此刻我们距离物理真实的图像编辑还有多远？为回答这个问题，我们引入PICABench，该基准系统性地评估物理真实感，涵盖八个子维度（涵盖光学、力学和状态转换），适用于大多数常见的编辑操作（添加、移除、属性更改等）。我们进一步提出了PICAEval，一种可靠的评估协议，利用视觉语言模型作为评判者，并结合每种情况的区域级人工标注和问题。除了基准测试外，我们还通过从视频中学习物理规律，构建了一个训练数据集PICA-100K。在评估了大多数主流模型后，我们发现物理真实感仍然是一个具有广阔探索空间的挑战性问题。我们希望我们的基准和提出的解决方案能为未来工作奠定基础，从简单的内容编辑迈向物理一致的真实感。

英文摘要：

Image editing has achieved remarkable progress recently. Modern editing models could already follow complex instructions to manipulate the original content. However, beyond completing the editing instructions, the accompanying physical effects are the key to the generation realism. For example, removing an object should also remove its shadow, reflections, and interactions with nearby objects. Unfortunately, existing models and benchmarks mainly focus on instruction completion but overlook these physical effects. So, at this moment, how far are we from physically realistic image editing? To answer this, we introduce PICABench, which systematically evaluates physical realism across eight sub-dimension (spanning optics, mechanics, and state transitions) for most of the common editing operations (add, remove, attribute change, etc.). We further propose the PICAEval, a reliable evaluation protocol that uses VLM-as-a-judge with per-case, region-level human annotations and questions. Beyond benchmarking, we also explore effective solutions by learning physics from videos and construct a training dataset PICA-100K. After evaluating most of the mainstream models, we observe that physical realism remains a challenging problem with large rooms to explore. We hope that our benchmark and proposed solutions can serve as a foundation for future work moving from naive content editing toward physically consistent realism.

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。