Field-Localized Forgery Detection for Digital Identity Documents

TL;DR

FLiD is a lightweight, field-focused forgery detection framework for digital identity documents that outperforms existing methods in accuracy and efficiency by localizing and analyzing critical identity regions.

Contribution

The paper introduces FLiD, a novel approach that localizes and classifies key identity fields, significantly improving forgery detection performance and efficiency over full-document methods.

Findings

FLiD achieves high AUC scores of 0.880 for face and 0.954 for text fields.

FLiD reduces false acceptance rates with EERs of 18.05% and 11.61%.

FLiD outperforms general-purpose detectors while using 13x fewer parameters.

Abstract

Digital identity verification systems used in remote onboarding rely on document images to authenticate users, making them vulnerable to localized manipulations of key identity fields such as facial photographs and textual information. Existing forgery detection methods, developed primarily for natural-image forensics, show limited transferability to structured identity documents. We propose FLiD, a lightweight field-localized framework that targets critical identity regions rather than processing full-document images. A fine-tuned object detector first localizes face and text fields; a frozen MobileNetV3-Small backbone then extracts compact field-level embeddings, which are classified by lightweight neural network with only 191K trainable parameters. FLiD achieves AUC scores of 0.880 (face), 0.954 (text), and 0.923 (both-field attacks), with corresponding EERs of 18.05%, 11.61%, and 15.16%, representing absolute reductions of 29-35 percentage points over a full-document baseline trained from scratch. FLiD also consistently outperforms general-purpose manipulation detectors (TruFor, MMFusion, UniVAD) across all attack scenarios while requiring 13x fewer parameters and 21x fewer FLOPs