Praveenkottari/BD3-Dataset

This repository provides the BD3 dataset, containing over 3,965 annotated RGB images. It supports evaluating computer vision techniques for automatic defect identification to improve building inspections. A comprehensive study and comparison results simulate the identification and classification of defects in various real-world built environments.

40
/ 100
Emerging

This project provides a comprehensive collection of over 3,965 annotated images of building defects, like cracks, peeling, and algae. It helps civil engineers, building inspectors, and urban planners develop and test automated systems to quickly and accurately identify common issues in urban infrastructure. The dataset allows users to train machine learning models that can process images and classify defects.

Use this if you are developing AI models for automated building inspections and need a diverse, real-world dataset to train and validate your defect detection systems.

Not ideal if you are looking for tools to perform inspections directly, as this provides data for building those tools, not the tools themselves.

building-inspection infrastructure-maintenance urban-planning structural-engineering defect-identification
No License No Package No Dependents
Maintenance 10 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 15 / 25

How are scores calculated?

Stars

26

Forks

5

Language

Jupyter Notebook

License

Last pushed

Feb 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/Praveenkottari/BD3-Dataset"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.