Question 1

How does it know what to keep?

Accepted Answer

U²-Net Tiny is a salient-object detection model — it predicts which pixels are "the subject" and which are "background context". The crop box centres on the subject's bounding box.

Question 2

What aspect ratios are available?

Accepted Answer

Square (1:1), portrait (3:4 and 4:5), landscape (4:3, 3:2, 16:9), and a free custom ratio. Each preset can also output a fixed pixel size.

Question 3

What if there are multiple subjects?

Accepted Answer

The model returns one saliency map covering all the salient regions; the crop tries to include the largest contiguous region. For multi-subject photos with very different subjects, a manual crop in Crop Image gives more control.

Question 4

Will my photo upload?

Accepted Answer

No. The U²-Net Tiny model (~6 MB) downloads once and caches; saliency inference and cropping both run in your browser tab.

Question 5

How is this different from AI Profile Picture Maker?

Accepted Answer

AI Profile Picture detects faces specifically. Smart Auto-Crop works on any subject — products, pets, landscapes — using saliency rather than face detection.

Question 6

How long does Smart Auto-Crop take to process a file?

Accepted Answer

Most jobs finish in seconds. Speed scales with input size and with how many CPU cycles your browser tab has available — the engine runs in your browser, so it shares resources with whatever else you have open. For inputs near the 25 MB ceiling, expect anywhere from a few seconds to roughly a minute on a typical laptop. Closing other heavy tabs noticeably speeds things up.

Question 7

Why did Smart Auto-Crop reject my input?

Accepted Answer

Failures usually fall into one of three buckets: the input is in an unsupported format, the input is over the size cap, or the input is structurally malformed (a truncated download, a partial export, or a stream the engine does not recognise). The first two are easy to confirm — check that your file is one of JPG, PNG, and WebP and that it is below 25 MB. For the third, opening the file in its native viewer first is the fastest way to confirm the source is intact.

Question 8

Can I process multiple files at once with Smart Auto-Crop?

Accepted Answer

Smart Auto-Crop processes one input at a time by design — it keeps memory usage predictable on lower-end devices and makes results easier to verify. To handle a folder, run the tool once per file; the page stays loaded between runs and remembers your last-used settings, so the second run is essentially instant.

Question 9

How accessible is the Smart Auto-Crop interface?

Accepted Answer

Smart Auto-Crop uses native HTML controls wherever possible, which means keyboard navigation, focus rings, and screen-reader labels work the way the platform expects. The drop zone accepts files via the keyboard-accessible file picker as well as drag-and-drop, and result downloads use standard browser download flows. If you spot an accessibility gap, Favtoo treats it as a bug worth fixing.

Question 10

Will I notice a difference in the output from Smart Auto-Crop?

Accepted Answer

Smart Auto-Crop is built to preserve quality wherever the underlying image format allows it. Operations that are mathematically lossless (e.g. structural transformations, lossless re-encoding) round-trip with no perceptible change. Operations that involve a lossy codec inevitably introduce small artefacts at the byte level, but the defaults aim at the sweet spot where output looks or sounds the same to a normal viewer or listener while still being meaningfully smaller or faster than the input.

Question 11

Does Smart Auto-Crop match what professional tools produce?

Accepted Answer

Smart Auto-Crop is built on the U²-Net Tiny salient-object detector running on ONNX Runtime Web (WebAssembly), which is the same class of engine used by professional image editing and conversion pipelines. For deterministic operations, the output is byte-identical to what an equivalent CLI run would produce; for operations involving a codec or a model, the result is well within the range of what comparable tools generate. If you have a specific reference output you need to match, run a small test job first to confirm the configuration produces what you expect.

Question 12

Does Smart Auto-Crop require a browser extension or plug-in?

Accepted Answer

No installation is needed. Smart Auto-Crop runs as a normal web page, with no browser extension, no native helper, and no separate desktop client to download. That is partly a privacy choice — extensions can request broad permissions, while a regular page is sandboxed by default — and partly a convenience one: you can use Smart Auto-Crop on any computer you have temporary access to without leaving anything installed on it.

Question 13

Is it safe to use Smart Auto-Crop on confidential files?

Accepted Answer

Your file is processed inside your browser by the U²-Net Tiny salient-object detector running on ONNX Runtime Web (WebAssembly). The engine reads the file's bytes from your tab's memory, computes the result, and writes the result back into the tab. You can confirm what the page does by opening developer tools and watching the Network tab during a run — the requests you see are for the tool's static assets only.

Smart Auto-Crop — Subject-Aware Image Cropping Online

Coming Soon

Smart Auto-Crop — Subject-Aware Image Cropping Online

Coming Soon

Explore more Image Tools