The Task
The goal was to create fabric masks that accurately mimic human facial features for biometric testing. Standard masks typically cover only the front of the face (180°), but our dataset required full coverage, including the back and sides, so respondents could move naturally without compromising data integrity.
Key requirements included:
- Full head contour for multi-angle recognition
- High-quality, realistic prints with modular replaceable
- Material flexibility, including thinner fabrics similar to nylon hosiery for enhanced biometric testing accuracy
The Solution
3D Design and Production
We developed a 3D breakdown of the head structure to guide the supplier. This allowed:
- Integration of fabric with complete head coverage
- Customizable inserts for facial features to increase realism
- Post-production assembly that preserves the head’s natural shape
Material Selection and Printing
Different fabrics were tested, including lightweight, semi-transparent options, which better simulate skin and allow anti-spoofing testing. Prints were optimized for:
- Hyper-realistic textures and pigmentation, including micro-details like freckles or subtle wrinkles
- Multi-layered construction for durable, realistic results
Process Optimization
Even with a capable supplier, fabric mask production involves scaling challenges. We:
- Standardized instructions for assembly and quality control
- Monitored consistency across samples and materials
- Ensured flexibility to swap parts for live testing scenarios
| Stage | Input | Workflow Scope | Main Quality Checks |
|---|---|---|---|
| Project Setup | Client platform & task requirements | Integration, task flow design, access configuration | System connectivity / Task logic consistency |
| Participant Onboarding | Contributor pool | Recruitment, onboarding, instruction delivery | Participant diversity / Instruction clarity |
| Attack Execution | User devices, printed images, replay materials | Print & replay attacks, iterative submissions | Attack variability / Scenario realism |
| Behavior Tracking | Attack attempt data | Tracking attempts, repeat participation, outcome logging | Data completeness / Behavioral consistency |
| Validation & Analysis | Collected attack data | System scoring review, performance analysis | Result consistency / Attack success evaluation |
| Reporting & Iteration | Validated attack datasets | Weekly reporting, feedback loops, system improvement tracking | Trend accuracy / Continuous performance alignment |
1–2 weeks
Pilot & Setup
2–3 weeks
Participant Onboarding
ongoing
Attack Collection & Iteration
weekly, ongoing
Monitoring & Reporting
The Results
- Successfully produced initial samples for early-stage testing with full head coverage
- Created a scalable workflow for future production of hyper-realistic fabric masks
- Established expertise in multi-angle, realistic mask fabrication for biometric validation
Biometric spoofing resilience is built through repeated real-world attack attempts, not static datasets. System performance improves when diverse participants continuously test its limits under varied conditions.
- Lucy Mamedoff
- Data Collection Project Manager
