explaingit

curzyori/rm-bg

Analysis updated 2026-05-18

58JavaScriptAudience · generalComplexity · 2/5LicenseSetup · easy

TLDR

A browser based tool that removes photo backgrounds and strips Gemini or NotebookLM watermarks entirely on your device.

Mindmap

mindmap
  root((RM-BG))
    What it does
      Removes photo backgrounds
      Strips Gemini watermark
      Strips NotebookLM watermark
      Runs fully in browser
    Tech stack
      Vite
      ONNX model
      WebGPU
      WASM
    Use cases
      Clean up AI generated images
      Make transparent graphics
      Isolate product photos
    Audience
      Vibe coders
      Designers
      Privacy conscious users

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Remove the background from a photo entirely inside your browser without uploading it anywhere.

USE CASE 2

Strip the visible watermark that Gemini or NotebookLM adds to an exported image.

USE CASE 3

Create a transparent product photo for a listing or design project.

USE CASE 4

Deploy your own copy of the tool to a static host like Vercel.

What is it built with?

JavaScriptViteONNXWebGPUWASM

How does it compare?

curzyori/rm-bglolita-cybe/doubao-dola-watermark-helperruanyf/wechat-format
Stars585858
LanguageJavaScriptJavaScriptJavaScript
Last pushed2023-10-19
MaintenanceDormant
Setup difficultyeasyeasyeasy
Complexity2/52/52/5
Audiencegeneralgeneralwriter

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

The hosted site works instantly with no setup, local development just needs npm install.

Free to use for any purpose under MIT, but one dependency is AGPL-3.0 licensed so check that if you reuse the code elsewhere.

In plain English

RM-BG is a browser based tool that removes the background from a photo and also strips watermarks that Google's Gemini and NotebookLM add to images they generate. Everything runs on your own device inside the browser tab, so no photo is ever uploaded to a server, which the author highlights as a privacy benefit. The background removal uses an on device model called ISNet, run through the imgly background removal library, which relies on ONNX and can use either WebGPU or WASM depending on what the browser supports. For the watermark removal, the project built its own canvas based engine, one path for the visible mark Gemini adds to exported images and a separate path for the footer style watermark that NotebookLM adds. The tool is aimed at anyone who wants to clean up an AI generated image, make a transparent graphic, or pull a single product out of its background, without installing software or creating an account. The interface supports four languages, English, Indonesian, Chinese, and Japanese, with a one click switcher, plus a light and dark theme that is remembered between visits. There are no usage limits, no credits system, and no sign up required. Under the hood the site is built with Vite 8, using its newer rolldown bundler, and plain JavaScript with no front end framework. To try it, you can use the hosted version at rm-bg.curzy.dev, or clone the repository, run npm install and npm run dev to start a local copy, or npm run build followed by npm run preview to produce a static dist folder that can be deployed to any static hosting service, with a Vercel configuration already included. The code itself is released under the MIT license, but one dependency, imgly's background removal library, carries the stronger AGPL-3.0 license, so anyone reusing this project in another product should check what that implies for their own code. The author built this as the twentieth entry in a personal fifty project challenge.

Copy-paste prompts

Prompt 1
Help me set up RM-BG locally with npm install and npm run dev.
Prompt 2
Explain how the ISNet background removal model works inside a browser using ONNX.
Prompt 3
Show me how to add a new watermark removal method similar to the Gemini and NotebookLM ones in RM-BG.
Prompt 4
Walk me through building and deploying RM-BG's dist folder to Vercel.

Frequently asked questions

What is rm-bg?

A browser based tool that removes photo backgrounds and strips Gemini or NotebookLM watermarks entirely on your device.

What language is rm-bg written in?

Mainly JavaScript. The stack also includes JavaScript, Vite, ONNX.

What license does rm-bg use?

Free to use for any purpose under MIT, but one dependency is AGPL-3.0 licensed so check that if you reuse the code elsewhere.

How hard is rm-bg to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is rm-bg for?

Mainly general.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.