One image to a full social pack
Set the focal point once. Every destination you choose is cropped around that same point, reviewed at its real ratio, and packaged into a ZIP with a manifest that records where each size came from.
- 1Image
- 2Destinations
- 3Focal point
- 4Review
Start with one image
Use the largest version you have — the pack crops down to every destination from this one file.
Nothing is uploaded. The file is read by this tab only.
Runs on your device. Your image is decoded, cropped and encoded by this browser tab. It is never uploaded, and there is no image-processing service behind this page.
Why a focal point rather than a template
Most batch resizers centre-crop everything, which works until the subject is not in the centre — and it almost never is. A portrait shot with headroom becomes a chin in a square avatar and a forehead in a 4:1 banner.
Marking the point that must stay in frame solves all of them at once, because a focal point is meaningful at any ratio while a crop rectangle is not. The same two numbers drive a 1500 × 500 header, a 1080 × 1920 Reel and a 112 × 112 emote. The review step then catches the handful of destinations where one focal point is not enough, so you can adjust before downloading rather than after.
Getting a good pack
- 01
Start from the largest source you have
A 4000 px original can fill every destination. A 1080 px screenshot will be enlarged for a YouTube banner, and the review step will flag exactly which crops are affected.
- 02
Review before you download, not after
The review grid shows each destination at its true ratio. A face that survives a square avatar can be cut in half by a 4:1 banner — that is the crop to fix.
- 03
Keep the manifest
It records which size came from which source document and when it was checked, which is what makes a handover to a designer or a client defensible six months later.
Questions people actually ask
How can one image work for a 3:1 header and a 9:16 Reel?
Because the shared unit is a focal point, not a rectangle. You mark the point that must stay in frame, and each destination computes its own crop around it. Where the source cannot fill a ratio, the pack fills the remainder with a blurred copy or a solid colour rather than letterboxing with black.
What is in the ZIP?
One folder per platform, each file named deterministically as platform-asset-widthxheight, plus a manifest.txt listing every file with its dimensions, format, exact byte size, whether it is within the platform’s published limit, the source URL for that size and the date it was checked.
Does anything get uploaded?
No. The decode, every crop, every encode and the ZIP itself are all built by your browser. There is no server involved in producing the pack, which is also why it works offline once the page has loaded.
How many destinations can I select?
All of them, though a large pack takes longer because each file is rendered at full resolution and encoded one at a time to keep memory use low. Emote and badge destinations expand into their full size ladders automatically.