A Bag of Tricks for Defending Deep Learning-based Face Recognition Pipelines against Backdoor Attacks
cs.CV, cs.AI, cs.CR, cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/biubug6/Pytorch_Retinaface
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
The gist: The protection of mature, multi-stage Face Recognition Systems is crucial to system providers.
Terminology
Abstract
The protection of mature, multi-stage Face Recognition Systems is crucial to system providers. This technical report targets the threat model most relevant to industry practitioners: although model training is performed in-house, training data is sourced from possibly-compromised third parties. Given this setting, we present three practical defense "tricks" against data poisoning-based backdoor attacks. The first operates at training-time, detecting poisoned identities during training or prior to a pipeline's deployment. The last two operate at inference-time, filtering suspicious inputs before an answer can be sent back to a potential attacker. Each method is presented with their motivation, algorithms and computational impact, and evaluated on realistic face recognition workloads. Our goal is to provide actionable hardening techniques to system providers.
Sources
- Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
- Rethinking the Vulnerabilities of Face Recognition Systems:From a Practical Perspective
- Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
- Backdoor Attacks on Deep Learning Face Detection
- SoK: On the Survivability of Backdoor Attacks on Unconstrained Face Recognition Systems
- Universal adversarial perturbations
- Mask and Restore: Blind Backdoor Defense at Test Time with Masked Autoencoder
- Face.evoLVe: A High-Performance Face Recognition Library
- Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in Unconstrained Environments
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs