面向开放式AI生成图像伪造定位的渐进式决策方法
Progressive Decision-Making for Localizing Open-Ended AI-Generated Image Forgeries
查看机构详情
- School of Artificial Intelligence, University of Science and Technology Beijing（北京科技大学人工智能学院）
- Institute of Artificial Intelligence, University of Science and Technology Beijing（北京科技大学人工智能研究院）
- Key Laboratory of Intelligent Bionic Unmanned Systems, Ministry of Education, University of Science and Technology Beijing（北京科技大学智能仿生无人系统教育部重点实验室）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
该研究针对AI生成图像伪造定位难题，提出基于EG-Mamba的渐进式决策更新方法，通过自适应序列决策提升定位性能，在未见AI生成伪造上表现更优。
中文摘要 AI 辅助
AI生成图像伪造的逼真度不断提升，固定篡改模式已难以对其进行表征。随着生成模型持续演进，期望定位模型仅通过大规模训练数据穷尽学习所有可能的伪造外观是不现实的。不过，许多AI生成伪造仍会留下细微的取证痕迹，尽管这些线索通常较弱且在不同区域的可靠性不均。因此，稳健的定位不仅需要提取有用的取证痕迹，还需要从不完整且模糊的证据中做出可靠决策。本文中，我们超越静态单次预测，将最终伪造定位重新表述为自适应序列决策更新过程，其中定位图被视为中间状态而非固定输出。我们的方法并非通过单次逐像素预测生成最终掩码，而是在可用证据、不确定性和边界条件的引导下渐进式更新定位状态。具体而言，我们首先通过轻量型决策证据投影器将介观痕迹转换为紧凑的决策证据，随后引入证据引导型Mamba（EG-Mamba）执行感知不确定性和边界的状态更新。该设计可保留可靠的篡改区域和背景区域，同时根据可用证据谨慎修正模糊区域。在传统和AI生成篡改基准上的大量实验验证了所提方法的有效性。值得注意的是，即便仅在传统篡改数据上训练，我们的方法在未见的AI生成伪造上仍带来了更大的性能提升，这表明渐进式决策更新对异质且难以穷尽学习的篡改痕迹尤为有用。
英文摘要
AI-generated image forgeries are becoming increasingly realistic and difficult to characterize with fixed manipulation patterns. As generative models continue to evolve, it is impractical to expect a localization model to exhaustively learn all possible forgery appearances from large-scale training data alone. Nevertheless, many AI-generated forgeries still leave subtle forensic traces, although these cues are often weak and unevenly reliable across regions. Therefore, robust localization requires not only extracting informative forensic traces, but also making reliable decisions from incomplete and ambiguous evidence. In this paper, we move beyond static one-shot prediction and reformulate final forgery localization as an adaptive sequential decision-updating process, where the localization map is treated as an intermediate state rather than a fixed output. Rather than producing the final mask via one-shot pixel-wise prediction, our method progressively updates the localization state guided by available evidence, uncertainty, and boundary conditions. Specifically, we first transform mesoscopic traces into compact decision evidence via a lightweight decision evidence projector, and then introduce Evidence-Guided Mamba (EG-Mamba) to perform uncertainty- and boundary-aware state updating. This design allows reliable manipulated and background regions to be preserved, while ambiguous regions are cautiously revised according to the available evidence. Extensive experiments on both conventional and AI-generated manipulation benchmarks validate the effectiveness of the proposed method. Notably, even when trained only on conventional manipulation data, our method brings larger gains on unseen AI-generated forgeries, indicating that progressive decision-updating is especially useful for heterogeneous and hard-to-exhaustively-learn manipulation traces.
