YOLOv8 Blueprint Detector
95% detection accuracy
Detects doors and windows in architectural blueprints. A YOLOv8 model fine-tuned on annotated drawings and served behind FastAPI, returning element coordinates, types and confidence scores as JSON for downstream take-off tools.
1 / 295% detection accuracy
Sub-second inference time
Real-time processing capability
SnapshotAt a glance
AI Engineer & Backend Developer
2 months
95% detection accuracy
Sub-second inference time
Real-time processing capability
Technologies Used
4 Technologies Integrated
Impact
Key Features
Feature Implementation
Project Vision
Streamline blueprint analysis with AI-powered detection.
Core Process
The process of Developing it.
Trained YOLOv8 model on architectural blueprints, built FastAPI backend, and created interactive frontend.
Build notesWhat I built
03Trained a custom YOLOv8 model specifically for architectural blueprint analysis, achieving 95% detection accuracy for doors and windows by creating an annotated dataset of architectural drawings and fine-tuning the model for precise element recognition, enabling sub-second inference times.
Built a FastAPI backend service that processes blueprint images through the YOLOv8 model, generating structured JSON outputs with detected element coordinates, types, and confidence scores, wiring computer vision capabilities with RESTful API architecture.
Developed an interactive drag-and-drop frontend with real-time visualization capabilities, allowing users to upload blueprints and instantly see detected architectural elements overlaid on the original image, delivering results in moments rather than hours of manual analysis.
Need for automated architectural element detection in blueprints.
Features
- Door & window detection
- JSON inference output
- Interactive drag-and-drop frontend
- Image visualization
- JSON API
Challenges
- Training on architectural drawings
- Handling various blueprint formats
