How it was implemented

The model is trained on labeled images of e-waste components. The web app uses a lightweight Flask/JS integration for inference, while Flutter handles mobile capture and upload.

What I learned

• Training and evaluating YOLOv11 models for object detection.
• Image annotation workflows and dataset preparation for accuracy.
• Data handling between clients and services for efficient inference.
• Building a Python backend with Flask to serve model predictions.
• Using Firebase for storage and database needs across platforms.