About the Face Blur
Face Blur is a free online tool that automatically detects and blurs all faces in a photo — helping you protect people's privacy before sharing images publicly.
How It Works
Your image is uploaded to our server where OpenCV processes it. The image is
first converted to greyscale, then OpenCV's Haar cascade classifier
(haarcascade_frontalface_default.xml) slides a detection window across the image
at multiple scales to find face-like patterns based on Haar-like features trained on thousands
of face images.
Each detected face region is blurred with a Gaussian blur kernel. You can choose the blur strength:
- Light — 25×25 kernel; faces are recognisable but softened
- Medium — 51×51 kernel; faces are significantly obscured (default)
- Heavy — 99×99 kernel; faces are completely unrecognisable
The output is a JPEG with all detected faces blurred.
What Is sortout.app?
sortout.app is a growing collection of focused web tools, each designed to do exactly one thing well. Every tool is free, requires no account, and is built to load fast anywhere.
Privacy
Your files are sent to our server over HTTPS, processed immediately, and deleted — never stored or shared. See our full Privacy Policy.
Frequently Asked Questions
Is this tool free?
Yes, completely free with no usage limits.
Do I need to create an account?
No. There is no sign-up, no login, and no registration of any kind.
What if a face is not detected?
Haar cascades work best on front-facing, well-lit faces. Profile views, faces at steep angles, very small faces, or heavily occluded faces may be missed. For critical privacy needs, manually verify the output.
Can it detect faces in crowds?
Yes, the detector scans the full image at multiple scales and can detect multiple faces. Performance may be slower on images with many small faces.
What formats are supported?
JPEG, PNG, and WebP. Maximum file size is 20 MB. The output is always JPEG.
Contact
Questions or feedback? Open an issue at github.com/sortout-app/feedback or email hello@sortout.app.