摘要：背景篡改是一种实用但定义尚不充分的图像取证场景：被篡改的证据可能位于显著前景对象之外，而许多评测仍强调以对象为中心的复制—移动、拼接或通用合成编辑。我们推出 BG-REAL，这是一个面向背景篡改检测与定位、以公共真实数据为锚定的基准套件。当前版本基于 Open Images V7 实例分割数据源构建，涵盖 1,200 个源图组的 7,000 个处理样本，包括 6,000 个公共数据锚定样本和 1,000 个合成对照样本。BG-REAL 覆盖六类编辑方式、配对真实对照、源图组划分、掩码与泄漏质量保证、599 条人工辅助质量控制记录、三个已完成的外部基线测试（TruFor、MVSS-Net 和 HiFi-Net），以及五个随机种子的模型评估。除汇总准确率外，我们还采用配对真实对照诊断：在留出验证数据上固定阈值后，衡量基线将重新编码的真实图像误判为篡改图像的频率；假阳性率从 0.57（TruFor，为最低值）到 1.00（若干较弱或利用掩码信息的基线）不等。这表明，重编码伪影是各基线共同面临的捷径风险，而非某一模型特有的问题。该发布版本提供构建流水线、评估协议、可直接用于论文的图表及复现文档。我们将 BG-REAL 定位为通用图像篡改定位基准在背景篡改方向上的补充，而非一个完全由真实数据构成或通用用途的基准。

Abstract:Background manipulation is a practical but under-specified image-forensics setting: the manipulated evidence can sit outside the salient foreground object, while many evaluations emphasize object-centric copy-move, splicing, or generic synthetic edits. We introduce BG-REAL, a public real-data anchored benchmark package for background manipulation detection and localization. The current release is built from Open Images V7 instance-segmentation sources and contains 7,000 processed samples over 1,200 source groups, including 6,000 public-data anchored samples and 1,000 synthetic control samples. BG-REAL covers six edit families, matched authentic controls, source-group splits, mask and leakage QA, 599 human-assisted quality-control rows, three completed external baselines (TruFor, MVSS-Net, and HiFi-Net), and five-seed model evaluation. Beyond aggregate accuracy, we use matched-authentic-control diagnostics to measure how often baselines misclassify re-encoded authentic images as manipulated at a threshold fixed on held-out validation data; false-positive rates range from 0.57 (TruFor, the lowest) to 1.00 (several weak or mask-informed baselines), indicating that re-encoding artifacts are a shared shortcut risk across baselines rather than a problem specific to any one model. The release provides the construction pipeline, evaluation protocol, paper-ready figures, and reproduction documentation. We frame BG-REAL as a background-manipulation-focused complement to general image-manipulation-localization benchmarks, not as a fully real-only or general-purpose benchmark.