Blurry Faces

Blurry Faces

A Gradio app from frapochetti that, going by its name, blurs faces in pictures, with benchmark scripts for face recognition and Kornia. For people who want to try face anonymizing in a browser.

Made by frapochetti on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: Blurry Faces is a Gradio app from frapochetti. Going by its name, it blurs faces in pictures to hide who is in them. The repository holds a model folder, sample images and two benchmark scripts, one for face recognition and one for Kornia. The README is empty, so what the controls do is not described.

Who it suits: It suits people who want to hide faces in photos before sharing them, and developers who want a small face-handling demo to read.

Industry | Who uses it | How they use it | Example
Media | Editors and journalists | Anonymize people in a photo | Upload a street photo and look at the result
Privacy | Privacy-minded users | Hide faces before posting a picture | Try a group photo
Development | Computer vision developers | Study a small face blurring demo | Read app.py and the benchmark scripts

You can skip it if you need guaranteed anonymization, since the source states no accuracy or detection limits.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Your own computer, on Windows, macOS or Linux

Use case | Who | What they get back
Blur faces in a photo | Editors and users | A copy of the picture with faces hidden, if the app works as named
Try the sample images | Newcomers | A quick look at the result on the included pictures
Read the benchmarks | Developers | Scripts that compare face tooling

How to set it up:

• Nothing to install if you use the Space online
• Python, if you run a copy on your own computer
• The Python packages in requirements.txt
• The system packages in packages.txt, which Hugging Face installs with apt-get

1. Open the Space's page on Hugging Face to use it online.
2. To run a copy, install huggingface_hub and download the Space into a folder.
3. Move into that folder and install Gradio, then the listed packages.
4. Start the app with Python.

pip install -U huggingface_hub
hf download frapochetti/blurry-faces --repo-type space --local-dir blurry-faces
cd blurry-faces
pip install gradio
pip install -r requirements.txt
python app.py

How to use it: Open the Space and give it a picture with faces in it. The repository includes sample images to try first. The README gives no steps, so follow the page's own controls. On your own copy, start the app with this command and open the local page Gradio prints.

python app.py

Example: Open the Space online and run one of the sample images that shows people. You should get the same picture back with the faces blurred, though the source does not describe the output.

For newcomers and experts: If you are new, try a sample image first, then a photo of your own. Check the result by eye, because a missed face stays visible.

If you know the field, the repository pairs a model folder with a face recognition benchmark script and a Kornia benchmark script. That points to a face detection step and image processing with Kornia, but the README does not confirm either. It gives no accuracy or speed figures.

You give the app a picture
The app finds the faces in it
The faces are blurred
The page shows the edited picture

Good to know:

• License: apache-2
• Cost to run: none named. The online Space runs on Hugging Face.
• Data it reaches: the pictures you give it, and the model files in its repository.
• Not for: guaranteed privacy. Missed faces stay visible, so check every result before you share it.

On Hugging Face, 13 November 2026

• Likes: 16
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/frapochetti/blurry-faces
• Files: https://huggingface.co/spaces/frapochetti/blurry-faces/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.