IDRetracor: Towards Visual Forensics Against Malicious Face Swapping

TL;DR

This paper introduces IDRetracor, a novel visual forensic method that retraces original faces from deepfake-generated fakes, enabling more reliable face attribution beyond simple fake/real classification.

Contribution

The paper proposes a new face retracing task and a model with mapping-aware convolutions to dynamically recover original faces from fake images, advancing face forensics.

Findings

IDRetracor effectively retraces original faces from various face swapping methods.

The method outperforms existing detection approaches in face attribution accuracy.

Extensive experiments validate the robustness and reliability of the retracing performance.

Abstract

The face swapping technique based on deepfake methods poses significant social risks to personal identity security. While numerous deepfake detection methods have been proposed as countermeasures against malicious face swapping, they can only output binary labels (Fake/Real) for distinguishing fake content without reliable and traceable evidence. To achieve visual forensics and target face attribution, we propose a novel task named face retracing, which considers retracing the original target face from the given fake one via inverse mapping. Toward this goal, we propose an IDRetracor that can retrace arbitrary original target identities from fake faces generated by multiple face swapping methods. Specifically, we first adopt a mapping resolver to perceive the possible solution space of the original target face for the inverse mappings. Then, we propose mapping-aware convolutions to retrace the original target face from the fake one. Such convolutions contain multiple kernels that can be combined under the control of the mapping resolver to tackle different face swapping mappings dynamically. Extensive experiments demonstrate that the IDRetracor exhibits promising retracing performance from both quantitative and qualitative perspectives.