Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (12.6%) to scientific vocabulary
Keywords
Repository
Repository for the EXI.AI-Q App prototype.
Basic Info
- Host: GitHub
- Owner: stefanpietrusky
- License: apache-2.0
- Language: Python
- Default Branch: main
- Homepage: https://downchurch.studio/ki/
- Size: 16.8 MB
Statistics
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
EXI.AI-Q-V1 APP
This repository, developed by Stefan Pietrusky, deals with a prototype that circumvents the problem of computationally intensive image neural networks or computer vision pipelines by storing the image content as simple text metadata that an LLM refers to during communication. Depending on the image, appropriate questions are generated, which are answered by the user and evaluated by the app without consuming high computing power.
The metadata for JPEG and GIF files is implemented using the external tool ExifTool. Since PNG files do not recognize EXIF segments but store them in chunks, Pillow is used here.
EXI.AI-Q Structure
The structure of the current [V1] EXI.AI-Q app is shown below.
EXI.AI-Q working principle
Below is a short GIF showing the structure and function of the app.
Installing and running the application
- Clone this repository on your local computer:
bash git clone https://github.com/stefanpietrusky/EXI.AI-Q.git - Install the required dependencies:
bash pip install -r requirements.txt - Install Ollama and load a model for example Llama3.2 (3B). Alternatively, another model can be used.
bash ollama pull llama3.2 - If you encounter issues with meta responses or the non-deterministic outputs of the model, you will need to adjust the parameters (temperature, seed, and top_p) of the LLM using a modelfile. An example can be found in the repository.
- Install the ExifTool for your system.
- Install Python 3.10.11.
- Adjust the paths for ExifTool, the image directories, and the LLM designation in the code.
- Create a file for the metadata or use the template in the repository.
- Implement the metadata into the corresponding images.
bash python metadata.py - Start the EXI.AI-Q V1 app.
bash python app.py
Owner
- Login: stefanpietrusky
- Kind: user
- Repositories: 1
- Profile: https://github.com/stefanpietrusky
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you cite this repository, please use the following reference."
title: "EXI.AI-Q V1"
authors:
- family-names: "Pietrusky"
given-names: "Stefan"
orcid: "https://orcid.org/0009-0008-9739-5542"
version: "1.0.0"
date-released: "2025-07-26"
GitHub Events
Total
- Push event: 8
- Create event: 2
Last Year
- Push event: 8
- Create event: 2
Issues and Pull Requests
Last synced: 7 months ago
Dependencies
- Flask *
- Pillow *
- exiftool *