Blur People & Vehicles in Photos on Mac, Automatically

Some photos are easy to publish. The tricky ones are full of people who never agreed to be in your post — the crowd behind your friends at a street market, the car parked right in front of your café shot, the other kids in a school-event picture. AI Anonymize in OtterPix is built for exactly those: it detects every person and vehicle in the photo, covers each one completely, face and license plate included, and then lets you decide — one by one — who stays recognizable.

There is also a quieter point that matters more here than anywhere else: a tool whose job is to protect the people in your photos should not begin by looking at your photos. OtterPix runs both detection and covering on your Mac’s Apple silicon. The unedited original never touches a server, because nothing is uploaded in the first place.

When covering people and vehicles is the right call

A useful rule of thumb: the people you photographed on purpose can be asked — everyone else ended up in the frame by accident. Anonymizing is how you publish the moment without publishing the bystanders. Typical cases:

• Street and travel photography — the scene is the subject; the strangers in it are not.
• Event recaps and community posts — your friends said yes, the rest of the room didn’t.
• School, club and team photos — other people’s kids deserve extra caution, whether in a parents’ group chat or a public feed.
• Selling a car online — the listing needs your car sharp; the passers-by and the neighbor’s car in the background don’t have to come along.

How to blur people and vehicles in OtterPix

Detection is automatic; your judgment comes after. The whole loop looks like this:

1. Open OtterPix, click All Features in the sidebar, and choose AI Anonymize.
2. Drop your photo into the Input pane on the left.
3. Click Anonymize. On-device AI finds the people and vehicles in the shot, numbers each detection, and shows the covered result in the Output pane.
4. Pick a Method: Blur, Mosaic, Hex Mosaic, Solid Color or Rainbow Blocks. Strength sets how heavy Blur and the two mosaics are, Solid Color has its own Color and Opacity, and Radius widens the covered area around each detection.
5. Click any numbered box in the input to exempt that person or vehicle — the output recomposes instantly, with no re-run. Click again to cover it again. The Targets chips switch whole classes at once, such as every car.
6. Click Save (⌘S) and choose where the anonymized copy goes. Your original file stays untouched.

Five styles — which one to pick?

All five hide identity; they send different signals. Blur reads as photographic — softened, like a shallow depth of field — so the mood of the image stays intact. Mosaic and Hex Mosaic read as deliberate redaction: nobody mistakes those cells for an accident. Solid Color is the bluntest, a flat shape in any color you choose, and Rainbow Blocks trades the stern look for a playful one, handy for a lighthearted post where a black shape would feel heavy.

Two practical tips. First, be generous with Strength (or Opacity for Solid Color) — a timid effect can leave someone recognizable to people who know them, and you can judge the result live in the output pane. Second, stay consistent within a series: switching styles from one photo to the next turns the redaction itself into the story.

Keep yourself sharp, anonymize everyone else

The per-instance toggles are what make this a publishing tool rather than a blunt filter. Detection deliberately casts a wide net first — every person, every vehicle — and then hands you the guest list. You keep the subjects who agreed: yourself in the conference photo, your family at the trailhead, your own car in the listing. Whoever just wandered through the frame gets covered.

Because a toggle or a style change only recomposes the existing result instead of re-running detection, each click updates the preview immediately — auditioning “cover everyone except…” combinations costs nothing. And if you haven’t subscribed to Pro: the daily free run is counted per loaded photo, not per adjustment. Run once, toggle as much as you like, try all five styles, export — all of it is still that day’s single free use.

When you’d rather remove than cover

Sometimes anonymized isn’t what you want — you want gone. A photobomber right next to you, a trash bin in an otherwise clean scene, a stray water bottle on the table. That’s a different job, and it belongs to AI Eraser: paint over the object with a brush and the background fills in naturally, as if it had never been there. You can keep refining the same photo until the last trace is gone.

A simple way to decide: cover when the honest record matters (the person was there, the car was parked there), erase when the object is just visual noise. And before anything goes online, remember the photo’s invisible layer — EXIF metadata can carry the exact GPS coordinates of where you took the shot. OtterPix’s free Metadata tool strips that in a couple of clicks; the related guide below walks through it.