Item logo image for Cutout — cut objects out of images

Cutout — cut objects out of images

ExtensionTools8 users
Item media 5 (screenshot) for Cutout — cut objects out of images
Item media 1 (screenshot) for Cutout — cut objects out of images
Item media 2 (screenshot) for Cutout — cut objects out of images
Item media 3 (screenshot) for Cutout — cut objects out of images
Item media 4 (screenshot) for Cutout — cut objects out of images
Item media 5 (screenshot) for Cutout — cut objects out of images
Item media 1 (screenshot) for Cutout — cut objects out of images
Item media 1 (screenshot) for Cutout — cut objects out of images
Item media 2 (screenshot) for Cutout — cut objects out of images
Item media 3 (screenshot) for Cutout — cut objects out of images
Item media 4 (screenshot) for Cutout — cut objects out of images
Item media 5 (screenshot) for Cutout — cut objects out of images

Overview

Smart object cutout from any image right on the page: click an object, auto-select, brush, PNG in original quality.

Cutout lets you cut objects out of images right on the website where you found them — and gives you a PNG with a transparent background. Hover over an image and an icon appears in the corner. Click it, click the object, and the neural network selects the whole thing. Copy the result to the clipboard or save it as a PNG. FEATURES - One-click selection — the SAM 2.1 neural network understands object boundaries. As you hover, it previews exactly what will be selected. - Refinement — Alt+click removes what you don't need, drag a box to select an area, and Tab cycles through options: the whole object or just a part of it. - Main subject — automatic cutout without a single click. - Brush and eraser — add or remove areas by hand, with adjustable size and hardness. - Precise edges — after your click, the neural network makes an extra pass along the outline in high-resolution tiles, and the edge snaps to the real boundary in the photo. Stray specks and small holes are cleaned up automatically. - Multiple objects — lock them in one after another, subtract unwanted parts, invert the selection. - Lossless export — pixels are taken from the original file one-to-one; only the transparency changes. Crop the PNG to the object and add padding if you like. - Dark and light themes, keyboard shortcuts for every action. PRIVACY All processing happens on your computer. The extension never uploads your photos and collects no data about you. The only network request is downloading the neural network files from huggingface.co on first run; after that, the model loads from the browser cache and the extension works offline. REQUIREMENTS Chrome 116+. On first run, the model is downloaded (about 170–330 MB). Fastest with a graphics card (WebGPU); it also runs on the CPU, just more slowly.

Details

  • Version
    1.3.2
  • Updated
    September 22, 2026
  • Offered by
    vlddvr
  • Size
    6.72MiB
  • Languages
    6 languages
  • Developer
    Email
    vladlenlukin213.3@gmail.com
  • Non-trader
    This developer has not identified itself as a trader. For consumers in the European Union, please note that consumer rights do not apply to contracts between you and this developer.

Privacy

Manage extensions and learn how they're being used in your organization
The developer has disclosed that it will not collect or use your data. To learn more, see the developer’s privacy policy.

This developer declares that your data is

  • Not being sold to third parties, outside of the approved use cases
  • Not being used or transferred for purposes that are unrelated to the item's core functionality
  • Not being used or transferred to determine creditworthiness or for lending purposes

Support

For help with questions, suggestions, or problems, please open this page on your desktop browser

Google apps