Automatically Detect & Blur Faces in Video OnlineFaces
Launch VideoBGBlur pre-configured for multi-face detection and tracking. Protect bystander, interviewee, and minor identities in street interviews, vlogs, and bodycam footage without manual keyframing.
Auto-Track & Blur Faces · Gaussian
Auto-detect and track faces across all frames. Adjust Gaussian, Mosaic, or Box blur free for 7 days.
Supported input: MP4, MOV, WebM video files (up to 200 MB · cloud jobs about 60s · up to 4K 4096×4096)
Export & billing: H.264 MP4 with original audio · a single-target 3s preview is 3 credits, and held credits return if no target is found
01 / WHY CHOOSE THIS PRESET
3s preview validation, 7-day mask reuse & transparent billing
Test 3 Seconds Before Full Render
Cloud presets can preview the first 3 seconds: 3 credits for automatic cutout or one blur target, 6 credits for faces and plates.
0-Credit Restyling & Blur Tuning
Cloud masks and tracking paths are cached for 7 days—swap backgrounds or adjust blur radius and feathering at zero extra AI cost.
Cloud AI or 100% Free Local WebGL
Use cloud AI for natural backgrounds and privacy tracking, or switch to our 100% free browser-local WebGL chroma keyer and alpha inspector.
100% Automatic Credit Release
If a face, license plate, or screen blur job detects zero targets in your clip, reserved credits are released automatically.
02 / HOW TO USE · THREE-STEP WORKFLOW
Standardized workflow from source clip to verified export
Upload Footage with Faces preset active
Open the workspace with `target=faces` pre-selected and import your MP4, MOV, or WebM video.
Preview 3 Seconds & Set Blur Radius
Run a 3-second preview to verify face tracking across motion, and configure Gaussian or Box blur strength, expansion, and feathering.
Export Redacted MP4 with Audio
Download the composed H.264 MP4 with original audio intact. Re-adjust blur intensity anytime within 7 days for 0 extra credits.
03 / TYPICAL USE CASES
Where teams deploy this preset
Bystander Privacy in Public Filming
Automatically anonymize passersby in crowded city walks, documentaries, and street interviews.
Student & Patient Identity Protection
Blur minors and patients in classroom recordings or clinical training videos for GDPR/HIPAA compliance.
Bodycam & CCTV Evidence Redaction
Produce compliant evidentiary video releases without hours of manual mask tracking.
04 / FREQUENTLY ASKED QUESTIONS
Questions about this workflow
Can it track multiple moving faces in the same video?
Yes. The AI detector tracks multiple faces simultaneously across camera pans, subject movement, and partial turns.
What happens if the AI finds no faces in my clip?
If zero faces are detected, the job stops automatically and 100% of your reserved credits are refunded to your account.
Can I make the face blur stronger without paying again?
Yes. Adjusting blur algorithm (Gaussian vs. Box), blur radius, mask expansion, or feathering on a cached mask costs 0 extra AI credits.
Does blurring faces remove my video audio?
No. VideoBGBlur preserves the original audio track in the exported H.264 MP4 file.
How do I prevent face edges from leaking during fast motion?
Increase the Mask Expansion slider to widen the blur region around each detected face and add Feather pixels for a smooth transition.
What video formats are supported?
MP4, MOV, and WebM files up to 200 MB, about 60 seconds per cloud job, and 4096×4096 resolution.
05 / ALL 8 VERIFIED PRESETS (4×2 SYMMETRIC MATRIX)
Explore all 8 specialized video preset pages
Remove Video Background
Start a fresh AI background-removal project without restoring an older workspace.
Blur License Plates in Video
Open VideoBGBlur with license-plate detection selected.
Blur Screens in Video
Open VideoBGBlur with the verified screen target selected.
Blur Faces & License Plates
Auto-track and blur both faces and vehicle license plates in one outdoor pass.
Remove Green Screen from Video
Open the browser-local chroma-key tool with the green preset selected.
Remove Blue Screen from Video
Open the browser-local chroma-key tool with the blue preset selected.