https://github.com/kikugo/isitadinosaur
A Streamlit app using Google's Gemini models to analyze images.
https://github.com/kikugo/isitadinosaur
ai asyncio gemini generative-ai google-ai image-analysis image-processing machine-learning object-detection python streamlit web-app
Last synced: about 2 months ago
JSON representation
A Streamlit app using Google's Gemini models to analyze images.
- Host: GitHub
- URL: https://github.com/kikugo/isitadinosaur
- Owner: kikugo
- License: gpl-3.0
- Created: 2025-02-17T05:27:03.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-17T06:14:19.000Z (over 1 year ago)
- Last Synced: 2025-02-17T06:24:06.207Z (over 1 year ago)
- Topics: ai, asyncio, gemini, generative-ai, google-ai, image-analysis, image-processing, machine-learning, object-detection, python, streamlit, web-app
- Language: Python
- Homepage:
- Size: 0 Bytes
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
# ISITADINOSAUR
_Is it Jurassic or just a pic? Let our image checker decide!_
## What's this?!
This app takes an image as input and uses AI (Google's Gemini 2.0 Flash) to determine if a dinosaur is present, providing a humorous response. It's built with Python and Streamlit, and now features asynchronous processing for improved responsiveness!
## Features
* **Dinosaur Detection (Humorous):** The app analyzes your image and generates a funny response about whether a dinosaur appears to be present.
* **Multiple Prompt Styles:** Prompts are dynamically generated by AI (Gemini 2.0 Flash Lite), providing endless variety!
* **User-Selectable Dinosaur Personalities:** Choose a dinosaur personality (Grumpy T-Rex, Silly Stegosaurus, etc.) to influence the tone of the response.
* **Image-Based Prompt Modification:** The prompt is subtly adjusted based on the content of your image, making the responses more relevant.
* **Dinosaur Facts:** Learn a random dinosaur fact after each analysis!
* **User-Submitted Captions:** Add your own funny caption to the image.
* **Sound Effects (Optional):** A fun sound effect plays after the response (can be muted).
* **WebP Support:** Supports JPG, JPEG, PNG, and WebP image formats.
* **Serious Mode (Object Detection):** Switch to "Serious Mode" to identify the main objects in the image instead of getting a humorous dinosaur response.
* **Asynchronous Processing**: The app is now more smooth and responsive.
## Tech Stack
* Python
* Streamlit
* Google Gemini (2.0 Flash and 2.0 Flash Lite)
* asyncio
## How to Run (Locally)
1. **Clone the repository:**
```bash
git clone
cd ISITADINOSAUR
```
(Replace `` with the actual URL of your GitHub repository.)
2. **Install dependencies:**
```bash
pip install -r requirements.txt
```
3. **Create a `.env` file:**
* Create a file named `.env` in the project's root directory.
* Add your Google Gemini API key to the `.env` file:
```
GOOGLE_API_KEY=your_actual_api_key_here
```
(Replace `your_actual_api_key_here` with your key.)
4. **Run the app:**
```bash
streamlit run app.py
```
## Roadmap
* [ ] Explore LangChain integration.
* [x] Implement asynchronous processing for improved responsiveness.
* [x] Add "Serious Mode" for actual object detection.
* [x] Switch to Gemini 2.0 Flash and Flash Lite models.