Blur Face

The public page describes importing a photo, adjusting detected face regions, and exporting the result. It lists JPG, PNG, WebP, and browser-dependent HEIC support. The founder says it also handles batches of up to 20 photos, works on video, and runs entirely in the browser, so photos don't need to be uploaded to a server. Detection quality and export fidelity aren't shown on the public pages, so test it with your own photos before relying on it.

The founder built Blur Face after seeing a family member who teaches early childhood education blur children's faces by hand before sharing classroom photos. Doing that in a regular photo editor, face by face and photo by photo, is slow, and the aim was a browser tool with automatic detection, manual fixes and batch processing, with no account or install. In the founder's example, a teacher preparing 20 classroom photos runs them as one batch, reviews the detected faces, corrects any misses and downloads the edited set.

This may fit someone preparing ordinary photos for sharing who will inspect each face manually. The page warns that detection can miss faces and weak blur can leave recognizable structure. It advises reviewing the export at full size and considering a solid cover when recognition would have serious consequences. The founder is just as direct. Small faces, unusual angles, poor lighting and partly hidden faces can be missed, which matters most in group photos of children, and that's why every redaction can be added, removed or adjusted by hand. Longer or high-resolution videos are also slower on older devices.

The editable first pass is the useful idea, together with the warning that automated detection is only a starting point. The site says the tool is free, needs no signup, and processes source images on the device. Network behavior and redaction strength aren't something the public pages can confirm, so check a full-size export before sharing anything sensitive.