How to blur a face in InShot
InShot doesn't detect faces. You cover one by stacking an overlay clip, a mosaic effect and a mask, then keyframing the mask by hand as the person moves.
The overlay and mosaic method
The method is the same for photos and video; video just adds the keyframing step to track movement.
- Add your video, then tap Overlay (PIP) and add the same clip again on top. Line the overlay up so it sits directly over the original, frame for frame.
- Select the overlay clip and open Filter or Effects. Apply Mosaic for a pixelated cover or Blur for a softer one. This affects the whole overlay clip, not just the face.
- Tap Mask and choose a shape. A circle usually reads best over a face. Drag and resize it so it sits over just the face, leaving the rest of the overlay transparent.
- Feather the mask edge. Raise the feather value so the covered area blends into the frame instead of sitting on top of it like a sticker.
- Turn on keyframes and adjust the mask position wherever the face moves. Add a new keyframe at each point so InShot can interpolate the mask's position between them.
- Repeat the overlay-mosaic-mask setup for each additional face. Nothing carries over automatically between faces.
- Export. InShot's free tier adds a small watermark to exports; removing it needs the paid version.
Where this gets slow
As with most mobile editors that lack face detection, the work scales with how much the person moves and how many people are in frame. A static headshot is a quick mask; someone walking through a shot means checking and adjusting the mask position every second or so for the whole clip.
Multiple faces mean multiple overlay-mosaic-mask stacks running in parallel, each with its own keyframe track — and if you trim or re-time the clip afterwards, every one of those keyframes needs re-checking against the new timing.
The mask is also a plain shape, not a face outline, so a tight circle on a turning head either clips the edge or leaves a visible sliver uncovered — something that's easy to miss on a small phone screen and only shows up when you watch the export back at full size.
Re-editing afterwards has the same problem it does in most timeline-based apps: the mask's keyframes are tied to moments in time, not to the face itself, so trimming the clip or changing its speed after the blur is in place usually means redoing the keyframe track rather than just shifting it.
What tends to go wrong
A few specific issues come up with InShot's overlay-and-mask method:
- Overlay and original drift out of sync if the clips aren't trimmed to exactly the same length before you start masking.
- Doubled audio from the overlay clip, unless it's muted separately from the original.
- Mask feather set too low, which leaves a hard, obviously edited edge around the covered area.
- Multiple faces sharing one overlay instead of separate ones, which makes it impossible to keyframe them independently as they move apart.
InShot vs. Deface
| InShot | Deface | |
|---|---|---|
| Time for one still face | 2–3 minutes of setup | About half a second |
| Moving subject in video | Manual keyframe every second or so | Tracked automatically, frame by frame |
| Multiple faces | A new overlay, mask and keyframe track per face | Detected together, toggle each on or off |
| Works offline | Yes | Yes |
| Removes photo/video metadata | No | Yes — strips EXIF, GPS and timestamp on export |
| Cost | Free tier adds a watermark; removing it needs a subscription | Free to try, one-time Pro unlock removes the badge |
The faster way
Deface removes the overlay-mosaic-mask-keyframe chain entirely. Import the same photo or video and it detects every face on device in about half a second, tracking each one automatically through the whole shot — no mask to draw, no keyframes to set.
Pick blur, pixelate, a solid block or an emoji, and it's applied to every detected face at once, switchable afterwards without re-processing. Each face has its own on/off toggle, and a manual region tool handles anything the detector isn't meant to catch, like a sign or a screen. Export strips the file's metadata too, which InShot leaves in place.
Frequently asked questions
Does InShot detect faces automatically?
No. InShot has no face detection. Covering a face means manually masking a blur or mosaic effect over it and repositioning that mask by hand as the person moves.
Can InShot blur a moving face in a video?
Yes, but not automatically — mask a blur or mosaic effect to the face on an overlay clip, then add a keyframe every time the face moves enough to fall outside the mask.
How do I remove the InShot watermark after blurring a face?
The watermark is tied to InShot's free tier generally, not to the blur effect specifically — it's removed by upgrading to the paid version, the same as any other export.
Can I blur a face in an InShot photo instead of video?
Yes, and it's simpler — the same overlay, mosaic and mask steps apply, just without keyframing since there's no movement to track.
Is there a faster way to blur a face on iPhone than InShot?
Yes. Deface detects and tracks every face automatically, with no overlay or mask to build — pick a style and export.
Does re-trimming an InShot clip break an existing face blur?
It can. The mask's keyframes are tied to time, not to the face, so trimming the clip, changing its speed, or re-cutting it after the blur is set up usually shifts the mask out of alignment and needs re-keying.
Blur every face in one tap.
Deface finds the faces automatically, on your iPhone. Nothing is uploaded, nothing is watermarked.
