Face redaction
Face redaction (also searched as facial redaction) is covering a person’s head in a video so the released file cannot be used to identify them from that image. Video Blackout is face blurring software for business in the sense that it runs on a licensed Windows PC, not as a social-media filter. The cover is mosaic, blur, or a solid fill, burned into every exported frame.
Definition: auto-detect plus manual boxes
Two modes exist for the same region type. Auto-detect runs an on-device face detector and proposes boxes. You keep a person clear (interview subject, officer, teacher) or mark them for redaction. Manual boxes are drawn when the detector misses a profile, a child at the edge of frame, a reflection, or a helmet. Tracking can follow either a detected box or a drawn box so a walk down a corridor is not hundreds of keyframes.
Auto-redact faces and auto-blur faces, in this product, mean that first pass — not an unattended export. A detector does not know your exemption list.
Video Blackout capabilities
| Identifier | Auto-detect | Manual |
|---|---|---|
| Faces | Yes — on-device detector | Yes — draw or adjust boxes; keep or redact per person |
| License plates | Yes — on-device detector | Yes — draw or track boxes |
| Other on-screen PII | No | Yes — static or tracked regions (badges, screens, paperwork) |
| Audio | No — no speech recognition | Yes — operator marks time ranges; mute or beep on export |
What the operator actually does
- Open a local file. Nothing is uploaded to videoredaction.io or to a vendor cloud.
- Run detection. Deselect faces that must remain visible under policy.
- Draw extra boxes where detection failed. Track if the camera or the person moves.
- Choose mosaic, blur, or solid for the marked regions.
- Export hard-coded video. Keep the original; circulate only the redacted copy.
When detection misses (and why manual is first-class)
Real footage is messy: side profiles, hats, motion blur, crowded doorways, faces reflected in glass. Treating manual boxes as a “fallback apology” produces bad releases. In Video Blackout, a drawn box is the same region type as a detected one — resize it, track it, burn it. That is how schools finish a safeguarding clip and how records units finish a bodycam stop when the detector skips a passenger.
Effect types and permanence
Soft “privacy layers” in players can be turned off. Ours cannot: export rewrites the pixels before encode. Solid fill overwrites the region completely; mosaic and blur destroy the clear source in that area. Pick the style your counsel or DPIA expects and use it consistently across a program — not a different effect per operator whim.
Software to blur faces in video
“Software to blur faces in video,” “automatic face blurring,” and “facial redaction” are this page. Auto-detect proposes the boxes. You decide who stays visible. Tracking follows a person so you are not keyframing a corridor by hand. The how-to is how to redact faces in video. What the released face is called: what a redacted face is.
“How to blur a face in a picture” is a still-image job for a photo editor. This app opens video, not a one-off portrait workflow, and it does not run in the browser.
Offline face redaction
Detection and export run on the Windows PC that opened the file. After activation, teams can work without sending media to the public internet. That is the difference from consumer “blur face online” tools. Residency detail: offline video redaction. Detection assistance in product language: AI video redaction.
When faces are not enough
A blurred face next to a readable plate, a student ID, or a spoken name is still an identity. Continue with license plate blur (auto-detect plus manual boxes), PII video redaction for badges and screens, and audio redaction for manual mute/beep ranges. Bodycam context: bodycam redaction software. Schools and facilities: CCTV redaction tool. Research and ethics review: video anonymization. All capabilities: features overview.
FAQ
- Do you auto-redact faces without a human review?
- Detection proposes boxes on the PC. You confirm who stays visible and who is covered. Export is a human-approved, burned-in file.
- Can I draw a face box if detection misses someone?
- Yes. Manual boxes are first-class: draw, resize, and track. Auto-detect is an accelerator, not the only path.
- Is this face blurring software for business or a consumer app?
- Business and government desktop. There is no mobile “app to blur faces in video” that uploads to us.
- Can someone remove the blur from the exported file?
- No. Mosaic, blur, or solid replace the original pixels in a new encode. There is no hidden clear layer to peel off.
- What style should we choose — mosaic, blur, or solid?
- Policy decides. Solid destroys the region completely; mosaic and blur also remove clear source pixels. Pick one standard and apply it consistently.