Want to blur a face before an image goes online? Upload a photo, choose “Faces” and save the result. Detection runs automatically, even with many people in the picture. The service is free and usable without sign-up, and processing takes place on servers in the EU. If you like, you can blur license plates in the same pass.
Detecting faces automatically
Face detection uses the YuNet model, which Verpixle integrates through OpenCV (cv2.FaceDetectorYN). It is designed to find faces that are only a few pixels wide as well, such as passersby in the background or spectators in a crowd. Half-turned heads and partially covered faces with a cap, glasses or a mask are often detected too. With pure profile views, heavy motion blur and harsh backlighting, the model finds fewer faces.
Look at the result before you publish, and choose the “Strong” level for large, frontal faces.
How it works
- Select a file. Drag your photo into the box on the homepage. JPG, JPEG and PNG up to 200 MB are supported. You do not need an account.
- Tick “Faces”. Detection starts at the push of a button. Three levels (light, medium, strong) set how strongly the faces it finds are blurred.
- Check the result and save it. The anonymized image keeps its resolution. The uploaded file and the result disappear from the server automatically after no more than 24 hours.
When do faces have to be made unrecognizable?
There is no duty to make every face unrecognizable. It depends on whether a person is recognizable and whether you distribute the image. In Germany, § 22 KUG (Kunsturhebergesetz, the German act on the right to one's own image) generally requires the consent of the person portrayed before their likeness is distributed. § 23 KUG names exceptions, for example likenesses from the sphere of contemporary history, or shots in which people appear merely as incidental extras alongside a landscape or another location. Whether someone in the background is a mere incidental extra is decided case by case.
The GDPR applies alongside it across the EU: a recognizable face is personal data, and anyone who processes or publishes it needs a legal basis under Art. 6 GDPR, usually consent or a legitimate interest that outweighs the interests of the person portrayed. Anyone who makes other people's faces unrecognizable before an image goes into a blog, a shop, a marketplace or a social network usually does not have to make that assessment.
Note: This is not legal advice. For a binding assessment of your specific case, consult a lawyer.
Group photos, events and club photos
On a group photo, Verpixle makes every face it finds unrecognizable in a single pass, whether that is ten people lined up as a team or several dozen in a hall.
- For the club website, when not everyone in the picture has given consent (for minors, their legal guardians).
- For recaps, newsletters and social media posts after company parties, conferences and trade fairs where guests appear in the photos.
- For photos from schools, daycare and youth work. With children and teenagers, consent is harder to sort out; making faces unrecognizable is usually the simpler route.
- For vacation and street photos with passersby.
Blur or black out?
Verpixle softens the faces it finds with a Gaussian blur. Colors and rough outlines stay visible: you can tell that a person stood there, but not who. Blacking out replaces the area with an opaque bar. That is unambiguous but stands out more in the photo. Verpixle does not add bars.
What matters is the strength you choose. If it is too low, the eye region and the shape of the face can still be made out. For large, frontal faces “Strong” is the safe choice; for small heads in the background “Medium” is usually enough. Also, pass on only the anonymized version: sending the original as well makes the anonymization worthless.
Common questions
Is blurring faces free?
Yes. You upload a photo without an account and save the result free of charge. Free results carry a small watermark; on the paid Dealer plan it is dropped.
Does Verpixle detect small faces and profile shots too?
The YuNet model is designed for small faces and faces shot from the side, and it often finds people far back in the picture too. With pure profile views, heavy blur or harsh backlighting, a face can go undetected. So take a look at the result.
Does this also work on a group photo with many people?
Yes. Every detected face is blurred in a single pass, even with several dozen people in the picture.
Which files can I upload?
Images as JPG, JPEG or PNG. A file may be up to 200 MB.
Do I have to make other people's faces unrecognizable before publishing?
Under German law, § 22 KUG generally requires consent before someone's likeness is distributed, § 23 KUG sets out exceptions, and the processing needs a legal basis under Art. 6 GDPR. Anyone who blurs beforehand usually does not have to make that assessment at all. This is not legal advice.
Can I process faces and license plates together?
Yes. Before you start you choose whether faces, license plates or both are searched for – doing both together does not cost a second pass.
What happens to my uploaded files?
Processing takes place on servers in the EU. The uploaded file and the result are deleted automatically after no more than 24 hours and are not used for AI training.
Blur a face now
Select a file, tick “Faces”, save the result. No account and no installation.
Last updated: October 10, 2026