发表机构
KAIST（韩国科学技术院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对现有篡改定位方法在完全再生图像中失效的问题，提出半脆弱潜在空间扰动APT，通过锚点对齐特征实现篡改定位，在COCO数据集上FR IoU达0.92，性能优于基线，可泛化到未知篡改类型。
AI 中文摘要
主动篡改定位会在图像分发前嵌入不可感知的信号，以实现像素级篡改检测。现有方法假设拼接（SP）设置，即合成区域被复合到原始背景上，嵌入信号保持完整。但现实中基于扩散的修复在完全再生（FR）设置下运行，整个图像会经历去噪，破坏背景信号，导致现有框架失效。我们提出APT，一种半脆弱的潜在空间扰动，用于嵌入密集的逐向量定位信号。通过将每个空间特征向量对齐到固定锚点方向，APT在修复后通过合成前景与锚点对齐的背景特征之间的对齐差异来定位篡改。提出的难例挖掘损失和噪声扰动分支进一步强化了均匀对齐。在COCO数据集上的实验表明，APT的FR IoU达到0.92，优于最强基线WAM（0.84），而现有方法的性能接近随机（AUC为0.5），证明APT是一种可泛化到测试时未知篡改类型的实用取证框架。
英文摘要
Proactive tamper localization embeds an imperceptible signal into an image prior to distribution, enabling pixel-level manipulation detection. Existing methods assume a spliced (SP) setting, where synthesized regions are composited onto the original background, leaving embedded signals intact. However, real-world diffusion-based inpainting operates in a fully regenerated (FR) setting, where the entire image undergoes denoising, disrupting background signals and rendering existing frameworks ineffective. We propose APT, a semi-fragile latent-space perturbation that embeds a dense, vector-wise localization signal. By aligning each spatial feature vector toward a fixed anchor direction, APT localizes tampering via the alignment disparity between synthesized foreground and anchor-aligned background features after inpainting. The proposed hard negative mining loss and noisy perturbation branch further enforce uniform alignment. Experiments on COCO demonstrate that APT achieves an FR IoU of 0.92, outperforming the strongest baseline (WAM, 0.84), while existing methods collapse to near-random performance (AUC 0.5), establishing APT as a practical forensic framework generalizable across tampering types unknown at test time.
CommentsAccepted to ECCV 2026
