Construction tech · Computer vision
Blueprint Vision: Computer vision for construction takeoff and HVAC estimating
HVAC estimating starts with takeoff. Estimators count HVAC parts on construction drawings by hand. One building can take days. The symbols are small, dense, and different on each drawing set.
How it works
Find every symbol
A model draws a box around each symbol. Each box has a confidence score from 0 to 1.
A threshold decides which boxes stay. A high threshold removes wrong boxes. It also misses more real symbols.
Precision and recall move in opposite directions. The estimator must see both. My best detector reached mAP50 0.755.
Figure: a synthetic floor plan with detection boxes. Below it, a dot plot of confidence scores with a threshold line, and precision and recall curves that move in opposite directions as the threshold rises.
In the product
The estimator checks each AI suggestion
- The estimator clicks AI Assist. The model puts a box on each symbol that it finds.
- Each box shows a confidence score. A box with a low score can be wrong.
- The estimator rejects the wrong box. The app keeps this correction as training data.
- The estimator accepts the other boxes. The count changes from 12 to 15.
Simplified screen. The drawing, the scores and the counts are examples.


This video did not load.
45-second video
The correction flywheel
- A model makes a guess.
- A person corrects it.
- The fix becomes training data.
Transcript
- Estimators count HVAC parts on drawings by hand.
- The model draws a box on each part. Each box has a score.
- A threshold removes boxes with low scores.
- A person fixes the wrong boxes.
- Each fix becomes new training data.
- The model improves with each cycle.
- Data quality limited accuracy more than architecture.
The parts
- A detection pipeline (YOLO, then RF-DETR) trained on labeled blueprint tiles.
- An AI-assisted takeoff app: the estimator drives, AI suggests counts, duct sizes and symbol tags.
- A correction flywheel: every human fix becomes new training data.
Show 2 more
- Vector-text reading and tiled OCR to count grille tags from the PDF itself.
- Revision tools: slip-sheeting, FFT phase-correlation auto-align, added/removed diff.
By the numbers
- mAP50 0.755
- Best detector
- ~1,100
- Automated tests
- Auto-aligned
- Drawing revisions
What I learned
Data quality, not model architecture, limited accuracy. I changed the product to collect better data as people used it.
Need this for your team? Computer vision consulting