Spaces:
Sleeping
Sleeping
| license: mit | |
| title: Nano AI Image Analyzer | |
| sdk: streamlit | |
| emoji: π | |
| colorFrom: indigo | |
| colorTo: red | |
| pinned: false | |
| short_description: analyze image using gen ai | |
| # π± Nano AI Image Analyzer | |
| this is a Gen-AI project. that utilizes various AI models to analyze a picture. | |
| ### steps | |
| 1. create your virtual environment (venv) [optional] | |
| > Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser <br/> | |
| > .\venv\Scripts\Activate.ps1 #powershell <br/> | |
| > venv\Scripts\activate.bat #cmd <br/> | |
| 1. install all modules under `requirements.txt` using --> | |
| > pip install -r requirements.txt | |
| 2. run the app using | |
| > streamlit run app.py <br/> | |
| make sure to check `.streamlit/config.toml` for your streamlit settings | |
| ## features offered | |
| <h3>π Detect Objects</h3> | |
| detects person, car, watch, TV etc <br/> | |
| creates a bounded box around it <br /> | |
| provides a table with rows representing object type, box coordinates, score | |
| <h3>π Describe Image</h3> | |
| generates a very small caption | |
| <h3>π Generate Story</h3> | |
| creates a story based on caption | |
| <h3>π¬ Chat system</h3> | |
| ask about image | |
| ## models used π€π» | |
| why don't you read the `apps.py`? π | |
| ## file structure | |
| ``` | |
| |-- .streamlit | |
| | |-- config.toml --> streamlit configuration | |
| | | |
| |-- app.py --> main app | |
| |-- functions.py --> functions used in the app. (it is not a module) | |
| |-- instruction.txt --> draft of README.md | |
| |-- requirements.txt --> all the pakages under venv. generated using `pip freeze` | |
| |-- type2.py \ | |
| |-- type3.py --> some extra template i used while developing | |
| |-- type4.py / | |
| ``` | |
| ## link of streamlit cloud | |
| [will be added](#) |