← Hashito System Home 日本語 Tools Blog Posters
Style 10 18%
PreviewNo image yet
Drop an image here Open one with “Choose image”, or drop it onto this frame.
Pasting from the clipboard (Ctrl / ⌘ + V) works too.
Automatic detection is not perfect. Profiles, backs of heads and small faces are often missed. Drag on the image to add whatever is left.
No areas yet.
Press “Detect faces”, or drag on the image.
Load an image to begin. Everything runs inside this browser.
How to use: open an image → “Detect faces” → drag to add anything missed → “Export”. Style can be blur, mosaic or a solid black fill. Strength maps to the blur radius or the mosaic tile size, and Margin controls how much space is added around each detected face (hair and chin). Click a box to select it, then press Delete or use “Delete” in the list. Detection and editing run in your browser, and the image is never sent to a server. Exported images do not carry Exif data (capture time, GPS location and so on).
To inspect or remove the location and capture time themselves, use the Image EXIF Editor; to change the file format, use the Image Format Converter. Our free learning posters explain data and web concepts with diagrams.
Related tools: Sort Image Visualizer GIF Animator RAW Preview Extractor PixelPurge

Frequently asked questions

Are my images uploaded to a server?

No. Face detection, the blur or mosaic itself, and the export all run inside your browser. The image is never sent to a server. Only the detection model (about 250KB) and its runtime (WebAssembly, about 10MB) are downloaded from their sources (jsDelivr and Google's model hosting). That download happens once, when you press Detect, and later visits reuse the browser cache.

What if a face is not detected?

Drag on the image to add that area manually. Profiles, backs of heads, small faces and faces hidden behind masks or hats are often missed by automatic detection, so always check the result with your own eyes before exporting. Boxes added automatically can also be selected and deleted.

Is Exif data such as the capture time or GPS location kept?

No. The export re-encodes the picture from a browser canvas, so Exif (capture time, camera model, GPS coordinates and so on) is not carried over. When you are hiding faces, leaving the location attached would defeat much of the purpose, so this is intentional. If you want to inspect or edit Exif itself, use the Image EXIF Editor.

Should I use blur, mosaic or a solid fill?

A solid fill is the safest choice when the area must not be recoverable. Blur and mosaic look more natural, but at weak settings a person can still be recognised. Raising the strength slider increases the blur radius or the mosaic tile size. Zoom into the preview and confirm that the face shape and hairstyle no longer identify the person before you export.