Blur Faces in a Photo (Automatic)

Protect people’s privacy before you share

Photos of protests, school events, street scenes, workplaces and incidents often include people who never agreed to be online. Blurring their faces is good manners, often a legal requirement under data protection rules such as GDPR, and important for children’s safety. This blur faces tool detects faces for you and hides them with one click.

How it works

When you add a photo, a face detection model (MediaPipe, about 5 MB) runs in your browser and draws a box around each face it finds. The model downloads once, then works offline. Everything – detection, blurring and saving – happens on your device.

If no faces are found, the tool tells you and you can drag over any area to blur it by hand.

Effects

• Blur – a soft, natural-looking blur. Increase Strength for larger faces.
• Pixelate – the classic mosaic look used by news broadcasters.
• Solid box – a solid block of colour. The most secure option, because nothing of the original remains.

Shape: Oval fits faces neatly and looks less harsh; Rectangle is better for licence plates, documents and screens.

Strength controls how strong the blur or how large the pixels are. On high-resolution photos, faces are big in pixel terms, so use higher strength.

Adding and removing areas

• Drag anywhere on the photo to add a box – useful for faces the detector missed, plus licence plates, house numbers, name badges, ID cards, computer screens and children’s school logos.
• Tap a box to remove it, for example if you want to keep your own face visible.
• Clear all boxes starts again.

Check before you share

Automatic detection is good but not perfect. It can miss faces that are very small, turned sideways, partly hidden by hands or hats, or blurred by motion. Look over the whole photo – including reflections in windows and mirrors – before posting.

Also consider what else identifies people: distinctive tattoos, clothing with names, house fronts and car registration plates.

Is blur reversible?

Strong blur, pixelation and solid boxes in the saved photo cannot be “un-blurred”, because the original pixels are replaced. Very weak blur on large faces may still let people recognise someone, so use enough strength – or Solid box for sensitive cases such as witnesses and minors.

Metadata is removed too

The photo you download is a freshly encoded image that contains no EXIF or GPS data, so it doesn’t reveal where or when it was taken or which phone took it. To remove metadata from other photos without editing them, use EXIF Remover.

Common uses

• Social media posts from events, classrooms and sports clubs.
• Journalism and citizen reporting.
• Real-estate and street photos showing passers-by.
• Dashcam and doorbell images shared with neighbours or insurers.
• Research and case studies that must anonymise participants.
• Screenshots containing names and profile photos.

Related tools

Crop out bystanders entirely with Crop Image, reduce the file size with Compress Image, or convert iPhone photos first with HEIC to JPG.

Privacy

The point of this tool is privacy, so your photo never leaves your device – it is never uploaded or stored.