banana-peel 0.1.0

Remove Gemini AI watermarks from images using LaMa AI inpainting

🍌 Banana Peel

Remove Gemini AI watermarks from images using LaMa AI inpainting — seamlessly, locally, and in one command.

"Peel away those watermarks like a banana!"

Features

• AI-powered — Uses the LaMa inpainting model for seamless watermark removal
• Zero config — Model auto-downloads on first run (~198 MB, one-time)
• Batch processing — Process entire directories with parallel execution
• Metadata preservation — EXIF data is carried over to output files
• Supports PNG, JPG, JPEG, WebP, GIF, BMP, TIFF

Samples

Original	Watermark Removed

How It Works

Gemini adds a semi-transparent watermark logo to the bottom-right corner of generated images. Banana Peel:

1. Detects the watermark position using known mask patterns
2. Inpaints the area with the LaMa neural network, which "hallucinates" realistic content to fill the gap
3. Blends the result seamlessly using feathered alpha compositing

The AI model runs 100% locally via ONNX Runtime — no cloud, no API keys, no data leaves your machine.

Installation

From Source

git clone https://github.com/neocorp/banana-peel.git
cd banana-peel
pip install .

With Poetry (development)

git clone https://github.com/neocorp/banana-peel.git
cd banana-peel
poetry install

Usage

Single Image

banana-peel image.png
# Output: image_clean.png

# Overwrite original
banana-peel image.png --overwrite

Directory (Batch Processing)

banana-peel ./photos/

# Custom suffix
banana-peel ./photos/ --suffix "_nowm"

First Run

On the first run, Banana Peel will automatically download the AI model (~198 MB). This only happens once — subsequent runs are instant.

🍌 Banana Peel - First Run Setup
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Downloading AI model (198 MB)... this only happens once.

Downloading lama_fp32.onnx ━━━━━━━━━━━━━━━━━━━━━ 100% 198.4/198.4 MB

✓ Model ready! Processing your image...

Options

banana-peel <PATH> [OPTIONS]

Arguments:
  PATH           Image file or directory to process

Options:
  --overwrite    Overwrite original files
  --suffix TEXT  Custom suffix for output files [default: _clean]
  --version      Show version and exit
  --help         Show help message and exit

Detection Rules

Banana Peel automatically detects watermark size based on image dimensions:

Image Size	Watermark	Margins
> 1024×1024	96×96 px	64 px
≤ 1024×1024	48×48 px	32 px

The watermark is always positioned at the bottom-right corner.

Development

git clone https://github.com/neocorp/banana-peel.git
cd banana-peel
poetry install
poetry run banana-peel --help

Running Tests

poetry run pytest

Credits

• LaMa Model: Resolution-robust Large Mask Inpainting by Samsung AI
• ONNX Model: Based on the model from dinoBOLT/Gemini-Watermark-Remover
• Watermark Masks: From journey-ad/gemini-watermark-remover and allenk/GeminiWatermarkTool

License

MIT License — See LICENSE file for details.

Disclaimer

USE AT YOUR OWN RISK

This tool modifies image files. While designed to work reliably, unexpected results may occur. The author assumes no responsibility for any data loss or unintended modifications.

Limitations

• Only removes Gemini visible watermarks (the semi-transparent logo in bottom-right corner)
• Does not remove invisible/steganographic watermarks like SynthID
• Designed for Gemini's current watermark pattern (as of 2026)