gemini_watermark_remover

KO | EN

gitlite — search

#typescript #ai-agents #ai #dsh-plugin #deepseek-harness #open-source #cli #claude-code #codex #developer-tools #react #windows

gemini_watermark_remover

★ 2 Open GitHub ↗

you can remove any number of images' watermark using this script

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Gemini Watermark Remover

A Python script that automates the removal of the Gemini watermark from images using an AI inpainting model. This project was born from a love for automation and a desire to have clean, watermark-free images.

## About The Project

This script recursively scans a specified directory for images (`.jpg`, `.jpeg`, `.png`, `.webp`, `.bmp`, `.tiff`) and automatically removes the standard Gemini watermark from the bottom-right corner. It uses the `simple-lama-inpainting` library, which leverages a powerful inpainting model to intelligently fill the area where the watermark used to be, preserving the original image quality.

### Why?

-   **Automation:** Manually removing watermarks is tedious. This script does it in bulk, saving you time and effort.
-   **Clean Aesthetics:** For projects, presentations, or personal collections, you might want images without the distracting watermark.
-   **A Fun Project:** It's a great example of applying machine learning for a practical, everyday problem.

## Features

-   **Bulk Processing:** Recursively finds and processes all images in a folder and its subfolders.
-   **AI-Powered Inpainting:** Intelligently removes the watermark without leaving a blurry mess.
-   **Automatic Detection:** Calculates the approximate watermark location based on image dimensions.
-   **GPU Acceleration:** Automatically uses a CUDA-enabled GPU if available (`torch` is properly configured) for significantly faster processing.
-   **Cross-Platform:** Runs on any system where Python and the required libraries are supported.
-   **Overwrites Files:** The script overwrites the original image files with the watermark-removed version. **Please back up your original images before running this script if you want to keep them.**

## Getting Started

### Prerequisites

-   Python 3.x
-   A CUDA-compatible GPU is recommended for speed but not required (the script will fall back to CPU).

### Installation

1.  Clone the repository:
    ```sh
    git clone https://github.com/your-username/gemini_watermark_remover.git
    cd gemini_watermark_remover
    ```
2.  Install the required Python packages:
    ```sh
    pip install -r requirements.txt
    ```

**Note:** `torch` installation can be tricky. If you have a CUDA-enabled GPU, it is highly recommended to follow the official PyTorch installation guide for your specific system to get GPU support: [https://pytorch.org/get-started/locally/](https://pytorch.org/get-started/locally/)

### How to Use

1.  Run the script from your terminal:
    ```sh
    python main.py
    ```
2.  The script will prompt you to enter the path to the root folder containing your images.
    -   You can press `Enter` to use the default path defined in the script.
    -   Or, you can paste your own folder path and press `Enter`.
3.  The script will then scan for images and begin the removal process. It will print the status for each file.
4.  Once finished, it will display a summary of how many images were processed.

## Technical Details

The script identifies the watermark's position in the bottom-right corner. It assumes two standard sizes for the watermark based on whether the image's largest dimension is greater than 1024 pixels.

1.  A **mask** is created over the calculated watermark area with some extra padding.
2.  This mask is processed with `MaxFilter` and a Gaussian blur to create softer edges, which helps the inpainting blend more seamlessly.
3.  The `simple_lama` model then takes the original image and the mask, and "inpaints" (fills in) the masked area based on the surrounding pixels.
4.  The final, edited image is saved, overwriting the original.

## Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are **greatly appreciated**.

1.  Fork the Project
2.  Create your Feature Branch (`git checkout -b feature/AmazingFeature`)
3.  Commit your Changes (`git commit -m 'Add some AmazingFeature'`)
4.  Push to the Branch (`git push origin feature/AmazingFeature`)
5.  Open a Pull Request

## License

Distributed under the MIT License. See `LICENSE` for more information.

---

*Disclaimer: This tool is for educational purposes and personal use. Please respect the copyright and terms of service of any image generation service.*

gemini_watermark_remover

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README