OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Backscrub

Backscrub is a C++ Linux application that uses deep semantic segmentation to replace the background in camera video and publishes the processed feed through a V4L2 loopback virtual camera.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Backscrub — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Backscrub when you want background replacement at the virtual-camera level so the processed feed can be selected in video software rather than relying on a meeting app's own effect.

Who it is for

  • Linux users who want a replacement-background camera feed for video calls.
  • Developers or technically confident users able to build C++ software and configure V4L2 devices.

Documented uses

  • Separate a person from the camera scene using semantic segmentation.
  • Composite a wallpaper image behind the person.
  • Expose processed video as a virtual camera input for conferencing or capture applications.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. The documented build and usage instructions target Linux.

  • The workflow depends on the V4L2 loopback kernel module as well as the Backscrub application.
  • The README lists tested Ubuntu and dependency versions and examples of specific conferencing and capture software versions.
  • The project documents a CMake build path; its root-directory make-based build is marked deprecated.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Build and start a Backscrub virtual camera

Build the C++ project and launch it with a camera input, V4L2 virtual device and replacement wallpaper.

Before you start

  • Linux system with OpenCV development files, build tools, v4l2loopback-dkms and curl.
  • Git and CMake.
  • A camera device, a wallpaper image and permission to load the V4L2 loopback kernel module.

Steps

  1. Install the listed dependencies and clone the repository recursively.

  2. Create a build directory, enter it, then run cmake .. followed by the command below.

    Documented command
    make -j $(nproc || echo 4)

  3. Load the v4l2loopback module with a virtual device, as in the README example.

  4. Run the built backscrub executable with -c for the camera device, -v for the virtual device and -b for the wallpaper path.

Expected outcome

A virtual camera feed carrying the camera video with a replacement background, available to applications that can select that device.

Guide limitations

  • The README's tested configurations are Ubuntu 18.04.5 and 20.04 on x86-64 with specified dependency versions.
  • Some cameras may require the -f MJPG option for higher resolutions.
  • The listed conferencing and capture application versions are specific examples, not a compatibility guarantee.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 November 2026
Original website title
floe/backscrub · Official homepage title · retained in its source language
How we review sourcesSuggest a correction