Daniel Gatis (danielgatis) · Background removal

Rembg

Rembg is an open-source tool for removing image backgrounds, made by Daniel Gatis. It can be used as a command-line tool, a Python library, an HTTP server or a Docker container. The README lists several options: choice of model (including u2netp, SAM with point prompts, and custom ONNX models), mask-only output, alpha matting, and colour decontamination for soft edges. It also supports batch folder processing with a watch mode, and a pixel-stream mode intended for use with FFmpeg on video frames. Optional cloud backends (withoutBG and Atlas Cloud) are available, and CPU, NVIDIA CUDA and AMD ROCm installs are supported through ONNX Runtime. It needs Python 3.11 to below 3.14. The project is MIT-licensed with 25,379 stars; the latest release is v2.0.87 (6 November 2026). No benchmark results are given in the text.

Open the source ↗

Similar Vision & Image releases

  • InternLumina-U2

    Shanghai AI Laboratory · Omni-Visual Diffusion LLM · Week 36, 2026

  • ReDesign

    jintae-00 (GitHub) · Visual Understanding · Week 31, 2026

  • MODUS

    EPFL / Apple / U. Copenhagen / CUHK / U. Geneva / Lambda · Visual Understanding · Week 31, 2026

  • NVIDIA Nemotron 3.5 Super VL

    NVIDIA, via Amazon Web Services · Agentic VLM · Week 47, 2026

  • Mirage

    Cactus · On-device vision model · Week 46, 2026

  • P-H3-Image

    Pruna AI · Open image model · Week 46, 2026

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.