Four kinds of blur, and one case that is not blur at all
The pages that rank for this are checklists for next time. Clean the lens, take the case off, tap to focus, hold still, do not pinch to zoom. Every item on that list is true and every one of them is advice for a photograph you have not taken yet.
The photograph you came here about already exists, and the useful question about it is a different one: which kind of blur is in it. There are four, they have different causes, and they leave different marks on the file. You can separate them by looking at the picture, without knowing anything about what happened when it was taken.
That matters because the four are not equally recoverable. Two of them respond to an enhancer, one of them responds a little, and one of them does not respond at all, and the sentence that tells you which is yours is the one worth reading before you try anything. There is also a fifth case where the photograph is not blurry and the phone is showing you a stand-in, which is worth ruling out first because it takes seconds.
Rule out the one that is not blur
Start here, because it costs nothing and it explains a surprising number of cases. If your photos look soft in the library and then sharpen a moment after you open one, nothing is wrong with the photograph. The phone is holding a smaller copy and fetching the original when you ask for it.
That is what the Optimize Storage setting in iCloud Photos does. Apple describes it as keeping space-saving versions of your photos and videos on the device while the original full-resolution versions are stored in iCloud, and it works through the items you open least often first. So an old photo you have not touched in a long time is exactly the one most likely to be showing you a stand-in.
The test is to open the photo, give it a moment on a connection, and look again. If it resolves, you were never looking at the real file, and anything you do to the version on screen is work on a copy that is about to be replaced. If it stays soft after it has downloaded, the blur is in the photograph and the rest of this page applies.
- Soft in the grid, sharp after opening: a stand-in, not a blurry photo.
- Soft everywhere including after it loads: real blur, carry on below.
- Soft only after sending it to someone: that is compression in transit, not your camera.
Case one: motion blur, which smears in one direction
This is the one with a signature you cannot mistake once you know to look for it. Everything soft in the frame is soft along the same axis. A point of light becomes a short line rather than a disc, text drags sideways, and an edge that runs parallel to the movement stays sharper than an edge that crosses it.
It happens when the shutter stays open long enough for something to move during the exposure, which indoors and after dark is most of the time. Either the phone moved in your hand or the subject moved, and you can often tell which: if the background is smeared too, the camera moved, and if only the person is smeared, they did.
This is the least recoverable of the four, and the reason is worth stating plainly. The light from one point in the scene was spread across a row of pixels during the exposure, so each pixel holds a mixture of several different parts of the subject. Sharpening a mixture gives you a crisper mixture. Nothing in the file says which part of the smear came from where.
Case two: missed focus, where something else is sharp
The test for this one is to look away from your subject. Somewhere in the frame, usually in front of or behind the thing you wanted, there is a plane that is genuinely sharp: a shoulder, a doorframe, the pattern on a sofa, a leaf. Find it and the diagnosis is settled.
The camera focused somewhere, it just did not focus where you meant. That happens when the tap landed on the wrong place, or when focus locked and then the subject moved, or when the nearest high-contrast thing in the frame won. Face detection makes this rarer than it used to be and does not remove it.
Recovery depends on how far the miss went. A small miss leaves the subject slightly soft but still structurally present, and sharpening has something real to work with. A large miss leaves a face with no eyelashes, no pores and no edge to the iris, and those were never recorded, so anything that appears there afterwards was generated from what a model expects a face to look like. That is a plausible face rather than this person, which matters a great deal for a photograph of someone you know.
- A sharp plane somewhere else in the frame means missed focus, not camera shake.
- Slightly soft subject: worth running through an enhancer.
- No fine detail left in the subject at all: an enhancer will invent it rather than recover it.
Case three: digital zoom, with no sharp plane anywhere
Here nothing in the frame is sharp, and the softness has no direction to it. Look closely and you find the opposite of grain: smooth patches where texture should be, edges that look slightly blocky or waxy, and fine detail that has not been smeared so much as flattened out of existence.
That is an upscaling signature. Past the longest optical lens your phone has, pinching further does not change the lens, it crops the sensor and enlarges the crop. The enlargement is arithmetic rather than photography, and the image you saved is a small image that has already been stretched once.
This is the case where an enhancer genuinely earns its place, and it is worth being clear about why. The subject was in focus and the scene was adequately lit. What is missing is resolution, which is exactly the problem an upscaler is built for, and the fact that the file has already been enlarged once does not prevent a better enlargement from replacing it. Of the four, this is the one most likely to improve enough to be worth keeping.
Case four: low light, which is grain with soft edges
The signature here is a speckle. Fine coloured noise in the shadows, especially in flat areas like a wall or a sky, combined with edges that are soft but not directional and areas of skin or fabric that look oddly smooth next to the speckle.
Two things are happening and the second one surprises people. The sensor was pushed to a high sensitivity, which produces the noise, and then the phone removed a lot of that noise before writing the file, which also removes fine detail, because at that scale noise and texture look the same to the process doing the removing. So a low-light photo arrives both noisier and smoother than the scene was.
This responds partially. An enhancer can clean the remaining speckle and firm up the edges that survived, and the result is usually a noticeable improvement that is still clearly a low-light photograph. What it cannot do is return the texture the noise reduction already took out, because that was gone before the file was saved.
- Speckled shadows plus smooth midtones: low light, not motion.
- Soft without direction: not camera shake, which always picks an axis.
- Worth trying, with modest expectations. The detail removed before saving is not in the file.
What the four signatures look like side by side
The quickest way to run the whole diagnosis is to open the photo, zoom in to full size on an area that should have fine detail, and ask three questions in order.
Does the softness run in one direction? If yes, it is motion, and it is the hard case. If no, is anything else in the frame sharp? If yes, it is focus, and how much is recoverable depends on how far the miss went. If nothing is sharp, is there speckle in the shadows? Speckle points to low light, and an absence of speckle with smooth flat patches instead points to digital zoom.
That order matters because the cases are not mutually exclusive and the hardest one wins. A dark photo taken at full zoom with a shaky hand has all three problems, and the motion is the one that decides how much of it comes back, so there is no point optimising for the others.
- Directional smear: motion. Hardest case, and no tool reverses it.
- Sharp plane elsewhere: missed focus. Recoverable in proportion to how small the miss was.
- Flat, waxy, no speckle: digital zoom. The best case for an enhancer.
- Speckled shadows, smooth skin: low light. Partial improvement, honestly got.
Which two are worth running through Unblur Photo
Unblur Photo enlarges and sharpens, and sometimes restores a face. That makes it the right instrument for a photograph whose problem is resolution and the wrong instrument for a photograph whose problem is that the light was recorded in the wrong place.
So the honest split is this. The digital zoom case is worth running, and it is the one that most often produces a result people keep. The mild low-light case is worth running. A small focus miss is worth trying, with the understanding that you are judging whether the fine detail that comes back is this person rather than a person. Motion blur is not worth running, and a badly missed focus is not either, because in both cases the detail a sharpener needs was never written to the file.
The repair half of this subject has its own page on this site rather than being restated here, and so does the question of what an upscaler can and cannot add. This page is for working out which case you are in before you open any of them.
If the photo still matters and none of this helped
Two things are worth doing before you give up on an image you care about. First, find the original rather than the copy. A photo that went through a messaging app, a social feed or an email has been recompressed, and recompression adds its own softness on top of whatever was already there. The version in your own library is better than the version somebody sent back to you.
Second, check whether there is a burst. A phone often saves several frames when a photo is taken in motion, and one of them is frequently sharper than the one it chose to show you. That costs a few seconds of looking and it beats every tool on this page, because a sharp frame is real detail rather than reconstructed detail.
And if the subject is still in front of you, retake it. That sentence is unsatisfying and it is also the only answer that recovers information rather than approximating it.
More guides
Questions
Why are my iPhone photos blurry?
Almost always one of four things, and they look different from each other. The shutter stayed open while something moved, which smears in one direction. Focus landed somewhere other than your subject, which leaves another plane sharp. You pinched past the optical lenses, so the phone enlarged a crop. Or the light was low, which leaves speckle in the shadows and smoothed-out texture elsewhere.
Why do my photos look blurry until I open them?
That is not blur. With Optimize Storage switched on, iCloud Photos keeps space-saving versions on the device and the full-resolution originals in iCloud, starting with the items you open least often, so an older photo shows you a smaller copy until the original is fetched. Open it, wait a moment on a connection, and look again before trying to fix anything.
How can I tell motion blur from bad focus?
Look for a direction and look for a sharp plane. Motion blur runs along one axis, so a point of light becomes a short streak and edges crossing the movement are softer than edges parallel to it. Missed focus leaves something else in the frame genuinely sharp, usually just in front of or behind what you wanted.
Can a blurry iPhone photo be fixed?
It depends which kind you have. A photo softened by digital zoom usually improves a lot, because the subject was in focus and only resolution is missing. A mildly soft focus miss improves somewhat. Motion blur does not come back, because each pixel holds a mixture of several parts of the subject and nothing in the file records which part came from where.
Does zooming in make iPhone photos blurry?
Past the longest optical lens on your phone, yes. Beyond that point pinching does not switch lens, it crops the sensor and enlarges the crop, so the file you save is a small image that has already been stretched. The signature is flat, slightly waxy detail with no sharp plane anywhere in the frame.
Will an enhancer make a blurry photo sharp?
On the resolution cases, often yes. On the cases where detail was never recorded, what you get is plausible detail rather than recovered detail, which is a real distinction if the photograph is of someone you know. A face with no eyelashes and no edge to the iris in the file comes back with eyelashes that were generated, not uncovered.