Featured Project

Snakinator

Snake Species Detection Computer Vision Model

Snakinator
Completed
Status
2023
Year

Overview

A snake species detection computer vision model trained on 30 different snake species using YOLOv8. The model is deployed on a web and mobile application to help users identify different snake species. The application allows users to upload images, videos or youtube links of snakes, which are then processed by the model to provide real-time identification of snake species along with confidence levels and if the species cannot be identified with high confidence, just identifying whether snake exist or not using a different computer vision model.

Key Features

Snake Detection
30 Snake Species Detection
Web App for live inference or uploading a video/image or youtube URL
Mobile Application for quick inference by clicking pictures and getting snake information from wikipedia API

Technology Stack

Yolov8
Streamlit
HuggingFace Spaces
Python
FastApi
Flutter
Roboflow
Kaggle Notebook

Demo Video

Screenshots

Project Documents

Snakinator Presentation

Presentation which summarizes the workflow, dataset and confusion matrix

Interested in this project?

Feel free to explore the live demo, check out the source code, or get in touch to discuss similar projects.