Squirrel-Lens

[MHacks 15] An object detection chrome extension that detects CHONKY squirrels in your browser.

This project aims to find out chonky squirrels, which the UofM is known for, in images. This project uses YOLOv8 as the base model, fined tuned for 10 epochs with 220+ handpicked chonky squirrel images from various sources including UofM reddit, instagram, twitter, and 200+ non-chonky squirrel images from squirrel data dataset. The total dataset size is 1122 images after augmentation. Base model used for this project is yolov8s, and it achieved a mAP50-95 of 45.0.

Install

Clone the whole repository using git clone https://github.com/SoulflareRC/Squirrel-Lens.git

Requirements

• ultralytics
• flask
• opencv
• PyTorch
• gradio
• skimage

Usage

Use the extension

1. Run the api server with python main.py
2. Open chrome extension tab by typing chrome://extensions/ in your browser
3. Turn on the developer mode on the upper right
   
4. Find the extension folder in the repo, drag it into the extension tab
5. The extension is successfully installed when you see this
   

Only use inference

1. Run the gradio app with python gradio_ui.py
2. You should see two links, one starts with 127.0.0.1 ..., another ends with gradio.live. You can use the former on your own machine, and you can share the latter with your friend!
3. Enjoy!

Confusion matrix of the model: