White Background Adder for Photos

Marisol pours candles on Sundays and photographs them on the same beige linen her grandmother left her, because the linen looks warm and the candles look expensive against it. Last month she listed forty of them in one sitting. Thirty-eight sailed through. Two came back flagged: main image must be pure white background, please resubmit. Forty retakes she did not have time for, one linen backdrop she was not about to abandon, and a Tuesday morning spent wondering whether 'basically white' counts. It does not.

Amazon's image rules are blunt on this point. The hero image must sit on pure white — RGB 255, 255, 255 — with the product filling at least 85% of the frame, no props, no text, no lifestyle staging. eBay is politer about it but points the same direction for new-condition listings. Marketplaces want the search grid to look like a catalog, not a mood board, because a uniform background lets shoppers compare shapes and prices without the marketing getting in the way.

The good news is that you almost never need to reshoot. If the photo is competently lit and the product reads clearly against its background, flattening it onto pure white is a mechanical operation: isolate the product, drop it onto an empty white canvas at the right size, pad it to the marketplace's aspect ratio, and re-encode. That is exactly what this tool does, on your machine, in seconds, with presets that speak Amazon and eBay's dialect of pixels.

This guide walks through the workflow, the marketplace specs you are actually satisfying, the compositing math happening under the hood, worked examples with real dimensions, and the honest limits — shadows, glass, and the shots that no algorithm should be allowed to flatten.

The tool takes a product photo in any common format — JPG, PNG, WebP, HEIC if your browser can decode it — removes the existing background, composites the product onto a pure white canvas, and pads the result to the aspect ratio and dimensions you pick. Two presets cover most e-commerce work: a 2000 by 2000 square tuned for Amazon's zoom requirement, and a 1600-pixel preset aligned with eBay's recommendation. A custom mode lets you type your own canvas size and fill percentage when a storefront or print layout needs something odd.

Two things it will not do. It will not invent a studio light: if your original is dim, orange or shot at noon on a marble counter with specular highlights everywhere, the flattened result inherits all of that. Background replacement is not exposure correction. And it will not add fake shadows back in — we will get to why marketplaces tolerate real soft shadows but the letter of the rules says 255, and why this tool errs on the side of the letter.

What it does quietly, as a side effect of the re-encode, is strip every scrap of metadata from the output. GPS coordinates, camera serial numbers, the timestamp of your kitchen-table shoot session — none of it survives. Sellers rarely mourn this. If you are the kind of person who wants to see what a photo is carrying before it goes anywhere, the EXIF Viewer next door does exactly that.

The whole operation takes under a minute per photo once you have your settings dialed. The tool sits at the top of this page; the steps below assume you are following along with a real file.

1. 1

   Drop in your photo

   Drag a file onto the drop zone or click to browse. Multiple files queue up and process one at a time, which keeps memory usage flat even with large catalogs.

2. 2

   Pick a preset or a custom canvas

   Amazon 2000px gives you a 2000 by 2000 square; eBay 1600px gives a 1600 by 1600 square. Custom lets you set any pixel dimensions for storefronts, print catalogs or marketplace categories that demand a specific ratio.

3. 3

   Set the fill percentage

   The default 85% matches Amazon's rule that the product must occupy at least 85% of the frame. Drop to 75% if you want breathing room for secondary images where the rule no longer applies.

4. 4

   Check the removal preview

   The isolation step shows which pixels it plans to keep. Products with hard edges against an even background clean up in one pass; anything fuzzy gets a second look in the edge-tolerance control.

5. 5

   Flatten and pad

   One click composites the product onto pure white, scales it to your fill target, and centers it with equal margins. The math section below shows exactly how those margins are computed, because you should never trust a centering algorithm you cannot check.

6. 6

   Download the result

   The output arrives as a JPEG at quality 90 — the sweet spot for photographic content on marketplaces. If the file needs to hit a specific size cap for a listing feed, run it through the Compress Image to Target KB tool afterward.

Marketplace image rules read like bureaucracy until you realize every clause exists to keep the search grid legible. Pure white kills visual noise between listings. The 85% fill rule stops sellers from shipping a tiny product floating in whitespace to look premium. The pixel minimums exist because zoom is a conversion feature — shoppers who pinch-zoom buy more, and zoom needs real resolution to work with.

One nuance worth knowing: the pure-white rule applies to the main image. Secondary images can show the product in use, in packaging, at angles, with props, in a lifestyle setting. Smart sellers flatten the hero image for compliance and let slots two through nine carry the personality. The linen backdrop is not dead — it is just demoted to image four.

Once the background is isolated, every pixel gets an alpha value — how much of it is product versus background. Fully opaque product pixels carry alpha 1.0, fully transparent background pixels carry alpha 0.0, and the fuzzy boundary in between carries everything in between. Compositing onto white then runs a formula you could do with a pencil: for each color channel, output equals source times alpha plus white times one minus alpha.

Run the numbers on a stubborn edge pixel. Say a soft shadow leaves a gray pixel at value 128 with alpha 0.6. The output is 128 × 0.6 + 255 × 0.4, which is 76.8 + 102, landing at 179 — a light gray, noticeably not white. The same pixel at alpha 0.2 becomes 128 × 0.2 + 255 × 0.8 = 25.6 + 204 = 230, close enough to white that JPEG compression will swallow the difference. At alpha 0.0 the pixel is exactly 255. This is why edge quality matters more than any other variable in background removal: the formula is honest, but it cannot turn a half-shadow into a wall.

The tool finishes by clipping every pixel the classifier marked as background to exactly 255, 255, 255 — not 254, not 250, not the slightly-blue white of an overexposed window. Marketplaces and their automated checkers sample corner and edge regions, and 'white enough for the eye' is not always white enough for a bot counting pixel values.

Color profiles deserve one paragraph of paranoia. Canvas operations run in sRGB, and the W3C's sRGB specification is what every marketplace means when it says sRGB — no fancy Display P3 or Adobe RGB adventures. If your original photo carries a wide-gamut profile, the decode step brings it back to sRGB, and saturated product colors may shift a hair. Compare the before and after at full zoom before you push forty files. If you also need a format change, the PNG to JPG Converter flattens transparency with the same compositing math.

Marketplace canvases are square, and product photos rarely are. A 1400 by 1050 shot — classic 4:3 from any phone — becomes a 2000 by 2000 square through scale-and-pad: compute the scale that fits the product inside 85% of the canvas, apply it, then center the result with equal margins on all sides. Here is the full arithmetic for that exact case.

Notice which dimension drove the calculation. The product's width hit the 85% ceiling first, so height had room to spare — with a wide-format product it would be the reverse, and a tall skinny item like a floor lamp leaves enormous side margins. Scale-to-fit inside the fill target guarantees the constraint is respected in both dimensions without ever stretching the image, because stretching a product photo distorts the product, and distorted products generate distorted returns.

Real products cast real shadows, and shadows are gray, and gray is not 255. Amazon's rule text says pure white; in practice faint contact shadows under a product usually survive review, while anything softer than that is a coin flip that depends on which reviewer or which automated pass greets your listing. This tool's default is compliance over beauty: contact shadows are kept only if you raise the shadow tolerance, and the output tells you plainly when a shadow region was clipped.

Glass and reflective metal are the hard cases. A clear glass candle holder photographed on linen has almost no edge — its boundary is defined by the background showing through it, and when you remove that background you remove the product. Transparency-aware isolation helps, but the honest answer is that glass shot on a contrasting backdrop needs either a reshoot on pure white in the first place or five minutes of manual masking in an editor. No one-click tool on the internet fixes this, and any that claims to is selling you something.

• Products with hair, fur, fibers or lace — expect semi-transparent edges and plan on a tolerance pass
• Items shot on backgrounds close in color to the product itself — white ceramic on white marble is a classic failure
• Photos with heavy vignetting or graduated backgrounds — removal treats gradients as edges and produces ragged masks
• Reflective packaging that mirrors the room — the mirror keeps the room, and the room is not white
• Very low-resolution originals — the marketplace zoom requirement cannot be invented after the fact

The last limit is arithmetic, not judgment: zoom. If your original is 800 pixels on the longest side, no amount of padding creates the detail that Amazon's 1000-pixel zoom threshold wants to display. Padding changes canvas dimensions; it cannot conjure detail that the sensor never recorded. Shoot big, flatten once, and downscale if you need to.

The obvious crowd is marketplace sellers: Amazon FBA folks maintaining dozens of ASINs, eBay dealers listing estate finds, Etsy sellers who cross-post to stay platform-independent. The less obvious crowd is larger. Shopify and WooCommerce store owners who want their category grids to stop looking like a ransom note of mixed backgrounds. Café owners formatting menu boards and printed menus. Crafters doing wholesale linesheets that buyers expect in catalog format. Real-estate agents tidying listing collateral. Anyone whose photos need to sit on a wall of white and look like they were all shot in the same studio, which is precisely what a catalog is.

Photographers use it too, in a slightly inverted way: as a finishing step before delivery. A client pays for product shots; the contract says marketplace-ready; the last mile of every file is flatten, pad, and compress. Doing that mile in a browser instead of a batch action in desktop software is mostly about speed and about not opening three heavyweight applications to move one file forward. For the heavier workflow of shrinking files to a marketplace's upload cap, the Image Compressor handles the follow-on.

You can rent a studio, you can wrestle a desktop editor, or you can flatten in the browser. Each has a place; only one of them fits inside the ten minutes between shooting the product and listing it.

The pattern that works for most small sellers is a hybrid: shoot on the plainest backdrop you own with even light, flatten everything in the browser, and escalate to studio or manual masking only for the hero products that carry the revenue. The tail of a catalog does not need studio treatment; it needs compliance and consistency, which is a much cheaper thing to manufacture.

Background removal is exactly the kind of job that sketchy upload-based sites love: your product photos, unreleased and possibly unreleased-forever, shipped to a server that might compress them, keep them, train on them or hand them to whoever asks next. This tool takes the other road. Decoding, isolation, compositing and re-encoding all happen in the canvas and typed arrays of your own browser. No network request carries a pixel. Close the tab and every working buffer is garbage-collected into oblivion.

For sellers this is not just comfort — unreleased product photos are trade secrets in a quiet way, and a catalog that leaks before launch is a catalog a competitor can undercut. The same privacy logic runs through the whole image-tools shelf here, because the safest upload is the one that never happens.

A marketplace-ready pipeline is rarely a single step. Flatten first, then shrink, convert, or inspect: PNG to JPG Converter for transparency-to-JPEG flattening, Compress Image to Target KB when a listing feed enforces a hard size cap, and Image Compressor for the general case. Before any photo goes public, the EXIF Remover makes sure it is not announcing your home address along with your product.