Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces BG-REAL, a novel benchmark for detecting and localizing background manipulations in images, addressing the gap in existing evaluations that focus primarily on foreground object edits. Built from Open Images V7, BG-REAL includes 7,000 samples across various edit families and provides a comprehensive evaluation framework that highlights the limitations of current models in accurately classifying manipulated images. Key findings reveal that false-positive rates for baseline models can reach up to 100%, underscoring a critical vulnerability in image forensics related to re-encoding artifacts.
False positives in background manipulation detection can reach 100%, revealing a systemic flaw in current forensic models.
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.