Skip to content

Latest commit

 

History

History
74 lines (47 loc) · 2.06 KB

readme.md

File metadata and controls

74 lines (47 loc) · 2.06 KB

Image Dataset Builder

This project is a Streamlit application that renames images based on the objects identified within them using the Ollama Llama 3.2 Vision model. The application allows users to specify the number of objects to identify and renames the images accordingly.

Features

  • Identify and name objects in images using Ollama Llama 3.2 Vision.
  • Rename images based on identified objects.
  • User-friendly interface with Streamlit.
  • Customizable input and output directories.
  • Adjustable number of objects to identify.

Installation

Clone the Repository

  1. Clone the repository: 
    git clone <repository-url>
cd image-dataset-builder

Install Dependencies

  1. Install the required dependencies: 
    pip install -r requirements.txt

Install Ollama

Windows

  1. Download the Ollama installer from the official website.
  2. Run the installer and follow the on-screen instructions.

macOS

  1. Download the Ollama installer from the official website.
  2. Open the downloaded .dmg file and drag the Ollama app to your Applications folder.

Linux

Install with one command: curl -fsSL https://ollama.com/install.sh | sh

Usage

  1. Run the Streamlit application:

    streamlit run app1.py

  2. Open your web browser and go to http://localhost:8501.

  3. Enter the input directory containing the images you want to process.

  4. Enter the output directory where the renamed images will be saved.

  5. Use the slider to select the number of objects to identify in each image.

  6. Click the "Elabora Immagini" button to start processing the images.

Example

  1. Input Directory: ./input_images
  2. Output Directory: ./output_images
  3. Number of objects to identify: 3

The application will process each image in the input directory, identify the specified number of objects, and save the renamed images in the output directory.

License

This project is licensed under the MIT License. See the LICENSE file for details.