bgremove-tool 0.1.0

AI-powered background removal CLI tool - Remove image backgrounds from command line

BG Remove Tool - Command Line Interface

A powerful command-line interface (CLI) tool for AI-powered background removal from images. Process single images or entire directories with simple commands.

Author: MAMUN ([e-mail])

🚀 Quick Start

# Install from PyPI
pip install bgremove-tool

# Remove background from an image
bgremove photo.jpg

# Process multiple images in a directory
bgremove --input-dir ./photos --output-dir ./results

# Get help
bgremove --help

Note: Install with pip install bgremove-tool, then use the bgremove command. This is a command-line tool, not a Python library.

Features

• Remove backgrounds from single images or batch process directories
• Smart output file naming: "_transparent.png" suffix for most files, "_bgremoved.png" for files already containing "transparent"
• Always outputs PNG format for transparency support
• Support for multiple input formats (PNG, JPG, JPEG, BMP, TIFF, WEBP)
• Overwrite protection with optional force overwrite
• Error handling and informative messages

Installation

Option 1: Install as Python Package (Recommended)

1. Clone this repository

2. Install the package:

   pip install -e .
   # or if using uv:
   uv pip install -e .
   

   This will install the bgremove command globally.

Option 2: Run Directly (Cross-platform)

1. Clone this repository

2. Install dependencies:

   pip install -r requirements.txt
   # or if using uv:
   uv sync
   

3. Use the appropriate launcher for your OS:

   • Linux/macOS: ./bgremove.sh
   • Windows: bgremove.bat
   • Any OS with Python: python bgremove.py

Option 3: Direct Python Execution

python main.py [arguments]

Usage

Remove background from a single image

# If installed as package:
bgremove image.jpg

# Using launchers:
./bgremove.sh image.jpg          # Linux/macOS
bgremove.bat image.jpg           # Windows
python bgremove.py image.jpg     # Any OS

# Direct execution:
python main.py image.jpg

This will create image_transparent.png in the same directory.

Specify output file

bgremove image.jpg -o output.png

Overwrite existing output file

bgremove image.jpg --overwrite

Batch process a directory

bgremove --input-dir ./images --output-dir ./output

This will process all supported images in ./images and save results to ./output.

Get help

python main.py --help

Command Line Options

• input: Input image file path (required for single file processing)
• -o, --output: Output file or directory path
• --input-dir: Input directory containing images to process
• --output-dir: Output directory for processed images
• --overwrite: Overwrite existing output files

Supported Formats

• PNG
• JPG/JPEG
• BMP
• TIFF/TIF
• WEBP

Cross-Platform Launchers

This tool provides multiple ways to run it on different operating systems:

OS	Command	Description
Any OS	bgremove	If installed as Python package
Linux/macOS	./bgremove.sh	Shell script launcher
Windows	bgremove.bat	Batch file launcher
Any OS	python bgremove.py	Python launcher script
Any OS	python main.py	Direct Python execution

Requirements

Minimum Requirements (CPU)

• Python 3.14+
• Memory: At least 4GB RAM (8GB recommended for large images)
• Storage: ~200MB for AI models (downloaded automatically on first use)
• Processing: May take 10-30 seconds per image depending on hardware
• rembg[cpu] (includes ONNX Runtime for CPU inference)
• numpy
• pillow

GPU Requirements (Optional - Faster Processing)

• CUDA-compatible GPU (NVIDIA)
• Additional ~2GB VRAM
• PyTorch with CUDA support
• rembg[gpu] (includes GPU-accelerated ONNX Runtime)
• torch>=1.9.0
• torchvision>=0.10.0

⚠️ Performance Note: The AI model loads ~500MB into memory. First run downloads models and may be slow. Close other memory-intensive applications.

Distribution & Packaging

For Developers - Building the Package

Using Modern Python Packaging (Recommended)

# Install build tool
pip install build

# Build distributions
python -m build

# This creates:
# dist/bgremove-0.1.0.tar.gz (source distribution)
# dist/bgremove-0.1.0-py3-none-any.whl (wheel)

Upload to PyPI

# Install upload tool
pip install twine

# Upload to PyPI (requires account)
twine upload dist/*

For End Users - Installation Options

Option 1: Install from PyPI (when published)

# CPU version (default)
pip install bgremove

# GPU version (if you have CUDA-compatible GPU)
pip install bgremove[gpu]

bgremove image.jpg

Option 2: Install from Source

# Clone repository
git clone https://github.com/bgremove/bgremove.git
cd bgremove

# CPU version (default)
pip install -r requirements.txt
pip install -e .

# GPU version (optional, requires CUDA-compatible GPU)
pip install torch torchvision rembg[gpu]

# Use the command
bgremove image.jpg

Option 3: Standalone Usage (No Installation Required)

# Download/clone the repository
git clone https://github.com/bgremove/bgremove.git
cd bgremove

# Install dependencies
pip install -r requirements.txt

# Use appropriate launcher for your OS:
./bgremove.sh image.jpg          # Linux/macOS
bgremove.bat image.jpg           # Windows
python bgremove.py image.jpg     # Any OS with Python

Package Contents

• bgremove/: Main Python package
• bgremove.sh: Linux/macOS launcher script
• bgremove.bat: Windows launcher script
• bgremove.py: Universal Python launcher
• pyproject.toml: Modern Python packaging configuration
• setup.py: Traditional setuptools configuration
• requirements.txt: Dependencies list
• README.md: Documentation
• LICENSE: MIT license

License

Add your license here