Processing a folder of images

Switch to Batch at the top of the window. The images are listed on the left, the selected one is shown in the middle, and the batch settings are on the right. Add images, choose where they go, press Enhance remaining, and leave it. The button counts the images it will enhance.

If no folder has been chosen yet, the panel says so beside the button, with Choose destination... right there. Once chosen, the folder shows under Saving to, with Change beside it.

Building the queue

Open images, or drag several files onto the window at once. Items take whatever settings are showing when they are added, so set up the enhancement first if you want it to apply.

Tick the images you want to act on. More list actions above the list moves, removes or clears them. The queue survives closing the application, so a batch interrupted by a restart is still there.

Saved edits are picked up automatically

If you have saved an edit beside a photograph (same folder, same name, with .lsenhance in place of the photograph's own extension, so beach.jpg has beach.lsenhance), queueing that photograph carries the edit with it, and the status line tells you how many of the images you just added have one. All of it applies: the crop, the orientation, the settings, which faces you chose, and the areas you painted as protected.

Each image carries its own edit. There is deliberately no way to apply one set of protected areas to a whole folder, because protection is stored for each pixel of one particular photograph: painting over a mole on one face and then laying those same pixels over forty other pictures would protect whatever happened to be in that spot, which is worse than doing nothing.

Batch settings copies the settings of the image on screen with Apply settings to checked or Apply settings to all; each button counts the images it will change. Either one changes the strengths and the size, and leaves each image's crop, turn and protected areas alone.

Work you drew in the window goes with the image

A crop, a turn or flip, protected areas and faces you left out of face recovery belong to that photograph. The batch uses them as the window holds them, saved or not: you do not need to save an edit first.

That work is kept while Litesum Enhance is open. When you close it with work not yet saved, it asks, and Save edit writes an edit beside each photograph so the queue picks the work up next time. If a photograph has changed on disk since you drew on it, the batch leaves that image alone and says so, and the rest carry on.

Two things a queue cannot do that you can do at the window:

• If the photograph is not the one the edit was made on, the image is left alone rather than enhanced. At the window Litesum can ask you whether to use it anyway; nobody is watching a queue, so it does not guess. The report says which images were skipped and why. This is on purpose: a file enhanced without its protected areas looks finished, and you would have no way to see that the thing you painted over had been rebuilt after all.
• An edit saved before Litesum recorded a fingerprint of the photograph is applied. If it has protected areas, the item finishes with a warning saying so. The size and date still matched, but nothing can confirm it is the same picture. Saving that edit again at the window records a fingerprint and the warning stops.

Leaving it running

While a batch is running, the computer is asked not to go to sleep until it finishes, so a queue started in the evening is done by the morning rather than suspended part-way through it.

The screen is still allowed to turn off. Only the machine is kept awake, which is what the queue needs. The request ends as soon as the batch does, and both the asking and the releasing are noted in the log beside your settings.

If your computer is set up in a way that does not allow this, the batch runs anyway and the log says the request could not be made, so a long queue may be interrupted if the machine sleeps.

Stopping, and starting again

Stop ends the run and leaves everything it had not reached in the queue, marked Cancelled. Pressing Enhance remaining afterwards picks those up and carries on; nothing that already finished is done again, and no original is touched either way.

Which images a button acts on

• Enhance remaining (N) runs the images still to do: queued, stopped, or changed since their result. An image with a current result, made in the single view or saved by an earlier run, is left as it is, and the panel says so.
• Enhance checked (N) runs the images you have ticked, including any already done. Use it to redo an image on purpose. Earlier files are kept, and a name already taken gets a number.
• Retry failed (N) runs only the images that failed.
• Export displayed image, at the top, exports the result of the image on screen, made in the single view.

When something fails

An item that fails says why in its row and in the report, and the run carries on to the next image rather than stopping the whole queue.

Retry failed appears when there is something to retry, and runs only the images that failed. Everything that finished is left alone: its output file stays exactly as it was, and is not rewritten.

It is a button rather than something the queue does by itself, on purpose. Most batch failures are temporary: a file open in another application, a destination that ran out of room, a photograph still being written when the queue reached it. Those usually work second time. But a queue that retried its own failures would turn one genuinely unreadable file into two attempts, and a full disk into four hundred, so you read the report, fix what needs fixing, and press it yourself.

How long it will take

Once there is something in the queue, the panel says how long the images still to do will take:

About 25 minutes for 47 images, from previous runs on this computer.

That comes from how fast this computer has actually been on jobs it has already finished, measured end to end rather than model-only: the whole wait, because the whole wait is what you sit through. Changing the output size or the image type changes the figure, since both change how much work each photograph is.

Three things to know about it.

It says how many images it covers. If some of the files could not be read, the line adds 3 more have not been measured and are not in that figure rather than describing a smaller job than the one you have.

It says where the number came from. From previous runs on this computer is a figure with your own history behind it. A reference estimate until a job has run on this computer means this computer has not run anything comparable yet, and the guess is correspondingly rough. The first job you finish at those settings replaces it with a real one.

If there is nothing honest to say, there is no line. No placeholder, no "calculating", no number invented from nothing. The line is also left out entirely when the whole queue would finish inside half a minute, because a warning about a wait shorter than reading the warning is just noise.

Working out the sizes reads each file's header and decodes nothing, so a folder of two hundred photographs is added as quickly as it ever was and the sizes appear behind the rows. They are remembered, so reopening the application does not read them all again, unless a file has changed on disk since, in which case that one is read afresh.

Trying a few first

A folder of two hundred photographs is a long way to find out that the settings were wrong for them. Try a few first enhances three of the queued images and stops. Everything else stays exactly as it was (still waiting, not paused and not cancelled), so when you have looked at the three, Enhance remaining runs the rest and nothing is enhanced twice.

The three are chosen, not taken off the top. One is picked for each kind of file in the queue, the largest of its kind first, and then, within whichever kind is commonest, the most heavily compressed one and the most ordinary one. The panel lists what it chose and why:

hillside.jpg: the largest of the 197 JPG files, at 6000 × 4000 (the slowest, and the likeliest to meet a limit) party-03.jpg: the most heavily compressed JPG here, at 0.21 bytes a pixel, where enhancement has most to do scan.png: the only PNG in the queue, at 1200 × 900

Those sentences matter more than the choice does. Choosing is done from the file headers (the dimensions, the size on disk and the format), which costs no time at all, and means the application has not looked at a single photograph. It cannot know that half the folder was taken in a dark room, or that one image is a photograph of a document. You know both. So the reasons are there for you to disagree with: if the three it picked are not the three you would have picked, open whichever one you want on its own and enhance that instead. The queue is untouched either way.

The button can be pressed once more than three images are waiting. Below that there is nothing to choose, and trying a few would just be running the queue under another name.

Applying settings to everything

Apply settings to all, under Batch settings, copies the photo type, enlargement and all three strengths onto every queued item, and turns automatic settings off for them, because otherwise the queue would measure each image and discard what you just applied. The status line says exactly what was applied. Apply settings to checked does the same for the images you have ticked. An image already saved keeps its file; choose Enhance checked to redo it with the new settings.

An image with a saved edit keeps its crop, its turn, its protected areas and the faces you left out. The settings you apply replace the edit's own.

Leave it alone if you want per-image measurement instead. That is usually the better choice for a folder of holiday photographs, which is rarely uniform: some are clean, some are compressed phone shots, and one set of strengths applied to all of them over-processes half.

Where the results go

Destination folder: choose it beside the Enhance button, or under Export. Originals are never modified and never written to. If Litesum cannot save to the folder you chose, it says so beside the button and nothing runs.

Format: one format for the whole batch:

• PNG (the default) and TIFF are lossless.
• WebP and JPEG are smaller. JPEG is lossy.
• Keep each image's own format is available, and the interface warns you what it means: any JPEG in the queue comes back as a JPEG, re-encoded, which discards some of the detail the model just reconstructed.

The default is PNG deliberately. Silently re-encoding a photographer's JPEGs after twenty seconds of AI reconstruction is the sort of thing that is only discovered by comparing file sizes a month later.

Keep source folder structure recreates the folders your images came from under the destination, rather than putting a hundred images from twelve folders into one.

File names get a suffix, such as holiday_enhanced_4x.png, and collisions are avoided rather than overwritten.

While it runs

The controls sit under the picture, beside the progress bar, while the batch runs.

One image at a time. That is deliberate: the graphics card is the limit, so two at once mostly compete for the same hardware while doubling the memory needed.

• Pause finishes the current image and then holds. It never abandons one half-processed.
• Skip abandons the current image and moves to the next.
• Stop ends the run. Anything not yet reached is marked cancelled, and you still get a report of what did get done.

A failure never stops the queue. A corrupt file, an unsupported format, an image too large for the memory available: each is recorded against that item and the run continues. Coming back to find it stopped on the second of fifty images would defeat the purpose.

Every image shows its state in words: Queued, Enhancing, Saved, Failed or Cancelled, with the reason. An image enhanced in the Single image view shows Ready, then Exported once you export it, or Changed if its settings move on.

Memory

Before each image, Litesum Enhance estimates what the job needs against the memory actually free. Anything that would not fit is skipped with a reason rather than started and allowed to exhaust the machine half-way through a night's work.

Keeping every file under a size

Maximum file size, under Quality and metadata in the Export section, applies to the whole run. Litesum lowers the quality as far as it needs to for each image, and no further than the point where a photograph starts to suffer.

An image that cannot be made small enough finishes as a warning rather than quietly as done, with the reason against it, and the report names it. That matters more here than at the window: across fifty images the report is the only way to know which ones missed. Only JPEG and WebP can honour a limit. PNG and TIFF are lossless, so their size follows the picture. A batch saved as PNG or TIFF is told so once, before it starts. With Keep each image's own format, each PNG or TIFF says so against its own row.

Every finished image is read back

When an image has been written, Litesum Enhance opens it again and checks it is the picture that was asked for: the right size, the right format, and decodable from beginning to end. An image that fails is marked as failed with the reason, and the file is left where it is rather than deleted: it is yours, and it is also the evidence of what went wrong.

This is not fussiness. A queue left running overnight can fill the disk, lose a network drive, or meet an encoder that gives up part way, and all three leave a file behind. Coming back to a folder of fifty names, there is nothing to tell a finished image from one whose pixels ran out half way down, and the header of a half-written file still says the right size, so a quick look would not catch it either. Reading the whole file is what catches it.

What this does not do is compare the file's pixels with what was on screen. Writing a file legitimately changes them a little where a colour profile is involved, so a check demanding they match exactly would fail on every correct image. The promise is narrower and true: the file opens, and it is the picture you asked for at the size you asked for.

The report

A text file is written beside the results listing, for every image:

• the source and output file names
• the dimensions in and out, and the scale
• the photo type used, and whether it was chosen automatically
• the three strengths that actually ran
• what a saved edit did: the crop, and how many pixels were protected
• the size of the file after it was read back and checked, and whether a size limit was met
• the format written and how long it took
• any warning or failure, in plain language

It also records how much the AI actually contributed to each image, the same figure the window shows. Across fifty images that line is what tells you one of them came back as little more than a resize.

On an image with protected areas, that figure describes the enhancement, not the file on disk: the protection is put back afterwards, and the protected part of the file had no AI contribution at all. The report says so on the line underneath rather than leaving you to work it out.

It records what was done, not what was asked for; the two differ whenever automatic settings are on. It contains file names and settings only; no image content, ever.