Dynamic Face Masking
Automatic face blur for every bystander in the frame — so incident footage is shareable with police, legal, and compliance teams the moment it's recorded.
Incident footage has one face that matters. Dynamic Face Masking obscures every other face in the scene — automatically, so evidence is shareable the moment it's recorded.
What It Does
DICE Dynamic Face Masking analyzes the scene, finds every face in it and masks all of them except the face of the person who triggered the event. The subject of the incident stays identifiable for operators, investigators and responders, while everyone else in the frame, like customers, employees, patients and passersby, is hidden automatically. The result is footage built with privacy in mind from the moment it’s captured.
The masking follows each face as people move through the frame, so it stays in place when people turn, walk through crowds or cross paths, instead of relying on a fixed blur area. When footage is exported or shared with police, an insurer, corporate security or a compliance reviewer, no manual redaction is needed before it leaves the system. When full anonymity is needed and someone’s body or clothing could still identify them, DICE Dynamic Privacy Masking applies the same approach to the whole person.
Who Uses It
Share the Clip the Same Day. No Waiting on Redaction.
A shoplifting clip from a busy Saturday afternoon can include dozens of customers' faces. DICE face masking lets loss prevention share it with police and corporate right away, without exposing shoppers who had nothing to do with it.
- Every customer's face except the subject's is masked automatically
- Clips can be shared with police and corporate the same day
- No waiting on manual redaction and no delays
A Patient's Visit Is Private. Their Face Shouldn't End Up in an Evidence File.
An incident at a pharmacy counter or clinic entrance often happens around patients whose visit is itself private information. DICE automatic face masking keeps the subject identifiable while patients stay out of the evidence.
- Patients' faces are masked automatically, so only the subject is visible
- Helps support HIPAA-aligned footage handling without a manual step
- Footage can be shared with police and risk management right away
Protect Every Student on Camera While Keeping the Subject Clear.
Footage of a trespasser or a fight almost always includes students. DICE face masking protects every student in the frame while keeping the person who triggered the event identifiable for administrators and responders.
- Every student's face except the subject's is masked automatically
- Helps support FERPA-aligned incident documentation
- Footage can be shared with administrators and police without extra redaction
Respond to Records Requests Faster. The Redaction Is Already Done.
Agencies that release footage under public records laws can respond faster and for less when bystanders' faces are already hidden. With DICE, the redaction work that usually slows releases down happens when the footage is captured, not after.
- Footage is ready for records requests as soon as it's captured
- Covers transit platforms, bus stops and public agency buildings
- Eliminates redaction time and cost on every request
Consistent Bystander Privacy at Every Site. No Staff Training Needed.
Companies applying GDPR-level privacy standards across all their locations get automatic bystander masking in every exported clip, at every site, without relying on staff to use redaction tools correctly.
- Face masking applied automatically and consistently at every camera location
- Helps support GDPR-aligned data minimization without redacting every clip by hand
- No staff training on redaction tools needed
How It Works
Detection
An event fires and streams to DICE's 24/7 monitoring operators with the subject's face visible for live assessment.
Voice-Down
The operator issues a live audio voice-down over on-site speakers.
Dispatch
When warranted, the operator escalates to dispatch, including direct digital 911 delivery.
No Delay
Because bystander faces are masked automatically, the clip the operator acted on is the same clip you can hand to responding officers — no redaction delay.
FAQs
-
No. Every face except the event subject's is masked. The person for whom the event was detected remains identifiable to DICE operators during response and in exported footage afterward — which is exactly what investigators need.
-
Masking follows each face as people move through the frame rather than blurring a fixed area, so bystanders stay obscured as they walk, turn, and pass one another. No manual cleanup is required before footage leaves the system.
-
No. Dynamic Face Masking locates faces to obscure them — it does not match faces against a database or establish anyone's identity. It reduces the personal data in your footage; it doesn't create any.