Find the identifiers.
Combine face and plate detection with OCR and selected personal-information categories in visible text.
pii zeroPII ZERO / PLATFORM
Detect personal information, apply masks across frames and inspect the result. Bring a repeatable privacy step to the video your team already collects.
Combine face and plate detection with OCR and selected personal-information categories in visible text.
Apply blur or solid masks to selected regions, with tracking and interpolation across the clip.
Review processed footage, annotated previews and per-clip detection records before release.
VIDEO REDACTION COVERAGE
Faces and plates are visual targets. Names, contact details and identity numbers are text targets. Badges, labels, paperwork and screens are surfaces where those identifiers may appear.
Visual detection
Text categories · OCR required
Text categories · OCR required
Text categories · OCR required
Optional configuration · OCR required
Optional screen-text policy
Text categories describe the configured English-language detection scope, not independently validated performance for every category. Coverage depends on readability, enabled configuration and capture conditions. Extended categories need additional review.
Names refer to written names, not whole-person masking. Screen text refers to OCR regions inside detected devices, not entire screens. The video pipeline does not include dedicated detection for handwriting, signatures, QR/barcodes, tattoos or logos, or selective spoken-PII redaction.
PROCESSING CAPABILITIES
Enable faces, plates, personal text or screen text for a run. OCR checks upright text and quarter-turn orientations, with English-language entity analysis.
Link detections across frames, interpolate regions between sampled frames and retain tracks through short detection gaps.
Choose blur or solid fill. Configure detection thresholds, sampling interval, track persistence and mask margins for your footage.
Re-scan processed video with the same detectors to flag residual findings for review. Use independently labeled samples to measure recall and precision.
THE DELIVERABLE
Use the outputs to assess coverage, resolve review items and define how redaction fits into your existing workflow.
LET’S TALK VIDEO REDACTION
Let’s define the categories, output requirements and local processing setup for your pilot.
15-minute introduction. No footage upload required.