反事实图像生成基准测试
Benchmarking Counterfactual Image Generation
AI总结：
针对反事实图像生成缺乏统一基准对比的问题，研究提出整合现有模型的对比测试框架，拓展至新数据集与因果图，验证Hierarchical VAEs的优越性，并发布可扩展的Python工具包。
摘要：
生成式AI已彻底变革视觉内容编辑，使用户能够轻松修改图像和视频。然而，并非所有编辑的性质都相同。要在自然图像或医学成像等领域实现逼真编辑，修改过程必须遵循数据生成过程固有的因果关系，这类图像编辑属于反事实图像生成范畴。评估反事实图像生成的难度极大：它不仅缺乏可观测的真实标签，还要求满足因果约束条件。尽管目前已有多种反事实图像生成方法和评估指标，但仍缺乏在统一设置下的全面比较。我们提出了一个对比框架，用于对反事实图像生成方法开展全面基准测试。我们整合了所有已用于该任务的模型，并将其拓展应用到新的数据集和因果图上，实验证明Hierarchical VAEs（层级变分自编码器）在大多数数据集和评估指标上表现更优。我们的框架以易用的Python包形式实现，可扩展纳入更多SCMs（结构因果模型）、因果方法、生成模型和数据集，供学界在此基础上开展研究。代码地址：https://github.com/gulnazaki/counterfactual-benchmark。
英文摘要：
Generative AI has revolutionised visual content editing, empowering users to effortlessly modify images and videos. However, not all edits are equal. To perform realistic edits in domains such as natural image or medical imaging, modifications must respect causal relationships inherent to the data generation process. Such image editing falls into the counterfactual image generation regime. Evaluating counterfactual image generation is substantially complex: not only it lacks observable ground truths, but also requires adherence to causal constraints. Although several counterfactual image generation methods and evaluation metrics exist, a comprehensive comparison within a unified setting is lacking. We present a comparison framework to thoroughly benchmark counterfactual image generation methods. We integrate all models that have been used for the task at hand and expand them to novel datasets and causal graphs, demonstrating the superiority of Hierarchical VAEs across most datasets and metrics. Our framework is implemented in a user-friendly Python package that can be extended to incorporate additional SCMs, causal methods, generative models, and datasets for the community to build on. Code: https://github.com/gulnazaki/counterfactual-benchmark.
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。
