Comparison
AI retouching against manual retouching
Where each one wins, and the split that works in practice.
Comparison
Where each one wins, and the split that works in practice.
This is the question behind every other page in this section. Should the skin be edited by software or by a person.
The honest answer is that it depends on the frame, and the split is predictable.
| What matters | AI retouching | Manual retouching |
|---|---|---|
| Time for 30 portraits | Under an hour once the tool is trained | A full day |
| Consistency across a session | Very high. The same decision every frame | Depends on the retoucher |
| Skin texture at 100 per cent | Smoothed unless the strength is reduced | Preserved |
| Face shape | Can drift on tools with shape sliders | Unchanged unless asked |
| Fabric and hair edges | The weakest point | Handled case by case |
| Cost per image | Pennies after the subscription | Charged per photo |
| Where files sit | Cloud tools upload them | Stays on the machine |
Prices and features on this page were read from each vendor's own website on 25 September 2026. Vendors change plans often. Check the current page before you buy.
Volume. Three hundred frames from an event need consistency more than they need artistry.
Repetition. The same dust spot on every frame is a job for software.
Noise and lens faults. These are measurable problems with measurable answers.
Turnaround. A client who needs images the same evening gets them.
A campaign image that will be printed large and looked at closely.
Any frame where the fabric is the subject. Software reads texture as noise.
Faces that will be recognised. A drifting jaw line is a different person.
Anything with a hand near the face, because edges meet and the model guesses.
Mariya's verdict. I run culling, noise and dust through software. Skin, colour and shape stay manual. That split has not changed in two years of trying to move it.
Let the batch tool set exposure, colour and noise across the whole session.
Pull out the ten frames the client will actually use.
Finish those by hand at full size.
The client sees a consistent gallery and ten images that hold up in print. The cost sits between the two extremes.
Prices for retouching are on the services page, charged per photo rather than per hour.
That order also protects the deadline. The gallery can go out while the hero frames are still being finished.
Take one portrait with side light and visible skin texture.
Run it through a batch retoucher at default strength and open both versions at one hundred per cent.
Look at four places in this order: the pores on the cheek, the line of the jaw, the hairline against the background, and the fabric nearest the face.
If all four survive, the tool is good enough for client work at that setting.
In practice the pores go first, and they go on every tool in the category.
On thirty frames, a person is faster than setting up a batch and checking it.
On three hundred, the arithmetic reverses completely.
The reason is consistency rather than speed. A human retoucher drifts across a long session, and software does not.
A gallery where frame two hundred looks like frame two is worth more to a client than any single perfect image.
Decides which blemish is temporary and which is part of the person.
Keeps freckles, scars and lines that belong to a face.
Reads the garment and protects its texture, because that is what the client is selling.
Looks at the frame as a photograph rather than as a collection of regions.
Takes responsibility when the client says it does not look like them.
Cull first, so nothing is processed that will not be delivered.
Run noise reduction on the RAW files.
Set colour and exposure per lighting setup, not per frame.
Run a light automated skin pass across the gallery at reduced strength.
Pull the frames the client will actually use and finish those by hand.
Sharpen last, locally, on the finished files only.
Retouching on this site is charged per photo, not per hour, and the rate is on the services page.
Per photo pricing exists because the work is per photo. A gallery of eighty and a campaign hero are different jobs.
A batch tool subscription sits between twelve and thirty-five US dollars a month depending on what it covers.
Adding those up across a year is the calculation most photographers never do.
A photographer buys a tool, runs it at default, and delivers a gallery of plastic skin.
The client says nothing, because they cannot name what is wrong.
They simply do not book again, and the photographer concludes that the market is difficult.
Reducing the strength by half prevents most of that, and it costs nothing.
| Case | Answer | Why |
|---|---|---|
| Thirty frames or fewer | A person | Setting up a batch costs more than doing the work |
| Three hundred frames | Software, then a person on the best ten | Consistency beats perfection at volume |
| Fabric is the product | A person | Denoising reads weave as noise |
| Face will be printed large | A person | Pores and jaw line show at print size |
| Event gallery, next-day delivery | Software | The deadline is the constraint |
This page is written by Mariya Harley, a fashion, portrait and hotel photographer working from a studio in west London.
The tools described here are the ones that reach a working photographer's desk, and the failures described are the ones that reach a client's screen.
Prices and features are read from vendor pages rather than repeated from other articles. The date of that check is printed above each table.
Nothing here is sponsored, and there are no affiliate links. A tool appears because it does the job, or because people ask about it.
What is deliberately absent: benchmark scores copied from press releases, rankings that change with the season, and any claim that software can replace direction, light or responsibility for the result.
If a price here is out of date, the vendor page wins. Tell me and I will correct it.
Every price and feature on this page was read from the vendor's own website on 25 September 2026.
Plans change often, and a tool that was a one-off purchase last year may be a subscription now.
Check the current page before you buy, and treat the figures here as a starting point for comparison rather than a quote.
Nothing on this page is sponsored. No vendor paid for a place in the tables, and there are no affiliate links.
This page is written by Mariya Harley, a fashion, portrait and hotel photographer working from a studio in west London.
The tools described here are the ones that reach a working photographer's desk, and the failures described are the ones that reach a client's screen.
Prices and features are read from vendor pages rather than repeated from other articles. The date of that check is printed above each table.
Nothing here is sponsored, and there are no affiliate links. A tool appears because it does the job, or because people ask about it.
What is deliberately absent: benchmark scores copied from press releases, rankings that change with the season, and any claim that software can replace direction, light or responsibility for the result.
If a price here is out of date, the vendor page wins. Tell me and I will correct it.