Most of us come to retouching through skin. You learn frequency separation, you learn dodge and burn, you learn exactly where the line sits between polished and plastic. You get good at faces.
Then you look at the frame again and realize the face is maybe fifteen percent of it. The rest is fabric.
I’ve spent thirty years shooting people who are wearing something, and the older I get the more convinced I am that the wardrobe is doing at least half the work. When we put Bill Gates in a black turtleneck for that Business Week cover, the turtleneck was the concept. When we showed up at Mar-a-Lago with eight white suits, the suit was the picture. If the clothes carry that much of the shot, they deserve better than being an afterthought in post — and everything you know about retouching faces gets you about halfway there.
Clothing has its own logic, and it punishes the habits that portrait work rewards.
The wrinkle problem
The first note from a client is almost always the same: get rid of the wrinkles. It sounds like a cleanup task. It isn’t.
A garment’s wrinkles are doing two jobs at once. Some are genuinely accidental — the fold from the shipping box, the crease from the hanger, the pull where a clamp held the fabric back, the sit-line across the lap of a suit your subject wore in the car. Those should go. Others are structural. They describe how the fabric falls, where it gathers at a seam, how much weight the material carries. Remove those and the garment stops reading as fabric. It flattens into a shape with a color on it.
So the judgment call happens on nearly every frame: is this crease telling the viewer something about the material, or is it just damage from handling? Get it wrong in one direction and the jacket looks like it came out of a suitcase. Get it wrong in the other and it looks like a 3D render of a jacket.
That call can’t really be automated, which is why the AI tools are still weakest exactly here. They’re good at finding a wrinkle. They’re not good at knowing which wrinkles are load-bearing.
The invisibility standard
The second thing that separates garment work from skin work is what success looks like.
In portraiture, a certain amount of visible retouching is often the point. The subject wants the polish, the magazine wants the polish, and everyone involved understands the image has been worked on. With clothing — especially anything that touches advertising or a brand’s own channels — the opposite is true. If a viewer can tell the garment was retouched, the retouching has failed. Not because it looks bad, but because a viewer who senses the image has been worked on starts wondering what the real thing looks like.
It’s a strange standard to work to: the better the job, the less evidence of it. The instinct to improve has to be checked constantly against a simpler question — does this still look like the thing that was actually in the room?
This is also why texture is what most often goes wrong. Smoothing is seductive. But a knit that has lost its loops, a linen that has lost its slub, a denim that has lost its weave — these read as wrong to a viewer who couldn’t tell you why. They just don’t believe the picture. Same failure mode as over-smoothed skin, different material.
Consistency is a craft problem, not a skill problem
There’s a third thing that catches people, and it only shows up at volume.
Retouch one garment well and you’ve made a series of judgment calls: how far to reduce the fold shadow, how far to push the color toward the reference, how much texture to hold. Retouch a full campaign — or a catalog running into the hundreds — and every one of those calls has to be made the same way on every image, including the ones you finish at the end of a long week.
Drift is the enemy. A set where image 12 and image 340 have been handled differently doesn’t fail loudly. It just reads as slightly incoherent, and the brand looks less considered than its competitors. Holding one standard across a large body of work is less about skill on any single image than about having a process that keeps the standard from wandering. It’s the part photographers underestimate right up until they’re on image 200 at midnight.
The cheapest retouching happens on set
Here’s the part I’d put above all of it: nothing you do in Photoshop is as cheap as a steamer.
Sweating the small stuff is what makes or breaks a shot. A stylist who steams the garment, tapes the hem, pins the back, and pulls the shoulder line straight before you ever release the shutter will save you hours of post and produce a better result than any amount of cloning. My wife Fazia has been the stylist on my shoots for most of my career, and the difference between a shoot with a real stylist and one without isn’t subtle — it shows up in the invoice from the retoucher.
Light plays into it too. Fabric texture lives or dies on the angle of your key. A source raked slightly across a weave gives you texture you can protect in post; a flat frontal source gives you a garment with no information in it, and no retoucher can put that back. That’s a lighting decision, made in about four seconds, that determines what’s possible for the next four hours.
Get the crew right and the post gets shorter. That’s the whole trick.
When to hand it off
None of this is beyond a good photographer. It’s just a different craft than the one you were hired for, and it scales badly against your time.
An hour spent deciding which creases in a blazer are structural is an hour you didn’t spend shooting, pitching, or getting the next job. The client is paying you for the shooting. If you’re doing volume apparel or lookbook work regularly, the math usually favors building a relationship with a retoucher who has made those calls ten thousand times they’re fast where you’re slow, and they can hold one standard across four hundred images because that’s the entire job description.
If you shoot apparel regularly, it’s worth knowing what clothing retouching looks like when it’s done as a specialty rather than as an afterthought to the shoot. There the wardrobe work is small enough per frame that it’s worth owning, because you’re the only one who knows what the fabric looked like in the room.
