When sharing group photos, event photos, or street scenes, you may need to hide the faces of people who should not be identifiable. Editing every face manually can become tedious when many people are present.
Automatic face detection can reduce the work, but detection is never perfect. A safer workflow is to detect first, then review the whole image and add manual masks where needed.
Do not stop at automatic detection
Face detection works best on larger, front-facing faces. Tiny distant faces, side profiles, partially hidden faces, or heavily blurred faces can be missed, and background patterns can occasionally be detected by mistake.
Use automatic detection to reduce manual work, then inspect the entire image at a useful zoom level before saving.
- Lower the minimum face size when many faces are small
- Raise the minimum size or adjust sensitivity if false detections increase
- Scan the whole image before exporting
Choose the redaction style for the situation
Mosaic and blur preserve more of the original look, while solid fills or eye bars are useful when you want stronger visual redaction. Emoji or custom image overlays can also be appropriate in casual contexts.
Whichever style you choose, leaving a little padding around the face helps prevent identifying details at the edges from remaining visible.
- Mosaic: obscure with coarse pixels
- Blur: blend the redaction into the photo
- Solid fill or eye bar: make the redaction obvious
Check for personal information beyond faces
Even after faces are hidden, name tags, addresses, license plates, computer screens, or documents can still expose personal information. Manual masks can also be used on these non-face areas.
Photo files may also contain GPS coordinates, capture time, and other metadata. Visible redaction and metadata inspection are separate checks, and both can matter before sharing.
Step by step
- Select, drop, or paste the image you want to redact.
- Run face detection and review the detected masks.
- If small faces are missed, adjust the minimum face size or sensitivity and run detection again.
- Add manual masks for missed faces, name tags, or other sensitive areas, then choose mosaic, blur, fill, or another style.
- Review the entire image one more time, then save as JPEG, PNG, or WebP.
Face Redactor
Detect faces locally and redact them with pixelation, blur, fills, eye bars, emoji, or custom images.
Tips and limitations
- For crowded photos, zoom in and inspect the edges and background as well as the main subjects.
- Aggressive detection settings can increase false positives, so combining automatic detection with manual correction is usually easier.
- If GPS or capture-time metadata also matters, inspect the finished image with Photo Privacy Inspector before sharing.
Frequently asked questions
Will every face always be detected automatically?
No. Tiny, side-facing, obscured, or blurred faces can be missed. Always review the whole image before saving and add manual masks when needed.
Can I hide areas other than faces?
Yes. Manual masks can cover name tags, addresses, license plates, screens, or any other area you choose.
Is the image uploaded to a Browser Kitty server?
Images selected in Face Redactor, along with face detection and redaction, are processed in the browser and are not uploaded to a Browser Kitty backend.