m1-cis 0.0.2

Context Image Search Utility

[!WARNING] This package is in development

Context Image Search

This utility provides you with advanced context to image search.

How it works

1. Decomposes given text (context) into search querries using gemini-2.5-flash-lite
2. Searches the querries on curated list of cc licensed images using google images
3. Analyzes the images for visual match using ML
4. Returns the best matching images

Requiered variablas

1. GEMINI_API_KEY - google gemini API key
2. GOOGLE_API_KEY - google custom search api key
3. GOOGLE_CX programmable search engine id

to aquire visit:

• google gemini key
• google custom search api -> Credentials
• google custom search engine id

Demo

You can check out the provided demo in ./demo. To get started:

1. cd into demo dir
2. run python -m venv .venv
3. run source .venv/bin/activate or venv\Scripts\activate if on windows
4. run pip install -r requierments.txt
5. populate .env.example and rename to .env
6. run streamlit run demo.py

Example

context: Rheinmetall to Acquire German Naval Shipbuilder NVL

resulting images: