Skip to main content
Reciclaje AI Demo

​
Welcome to Reciclaje AI

Reciclaje AI is an intelligent waste detection and recycling classification system that uses state-of-the-art YOLO object detection to identify and categorize recyclable materials in real-time. Built with Python and YOLOv8, this system helps automate the process of sorting waste into proper recycling categories. The system can detect and classify five different waste categories:
  • Metal - Aluminum cans, tin containers, metal objects
  • Glass - Glass bottles, jars, containers
  • Plastic - Plastic bottles, containers, packaging
  • Carton - Cardboard boxes, paper cartons
  • Medical - Medical waste and supplies

​
Key Features

Real-Time Detection

Process live webcam feeds with instant waste classification and bounding box visualization

5 Waste Categories

Accurately identifies Metal, Glass, Plastic, Carton, and Medical waste

Multiple Interfaces

Choose between simple CLI or interactive GUI with visual feedback

Pre-Trained Model

Ready-to-use YOLOv8 model trained on diverse waste imagery, available on HuggingFace

Confidence Scoring

Get confidence percentages for each detection to assess accuracy

Educational Resource

Perfect for learning Python, AI, and computer vision fundamentals

​
Who Is This For?

Learn practical computer vision and object detection implementation using modern frameworks like YOLOv8 and OpenCV.
Hands-on project to understand AI/ML concepts, real-time inference, and environmental technology applications.
Ready-to-use educational tool to teach computer vision, environmental awareness, and Python programming.
Build recycling automation systems, smart waste bins, or environmental monitoring applications.

​
How It Works

Reciclaje AI uses the Ultralytics YOLOv8 model fine-tuned on waste imagery. The system:
  1. Captures frames from your webcam using OpenCV
  2. Processes each frame through the YOLO model
  3. Detects waste objects and classifies them into categories
  4. Draws bounding boxes with labels and confidence scores
  5. Displays the results in real-time
The model has been trained on a variety of waste images under different lighting conditions and perspectives. However, performance may vary with unusual lighting, image noise, or rare waste items not in the training set.

Get Started

Install and run your first detection in minutes

View Model

Download the pre-trained model from HuggingFace

How It Works

Understand the detection pipeline and architecture

GitHub Repository

View source code and contribute

​
Community & Support

Join the AprendeIngenia community to learn more about AI and Python:
New to computer vision? Check out the introductory video to understand the concepts before diving into the code.

Build docs developers (and LLMs) love

Get started for freeTalk to us