HN Showcase
Back to browse
QuantTakeoff – Construction PDFs to takeoff and 3D scene

QuantTakeoff – Construction PDFs to takeoff and 3D scene

by acaciabengo·May 16, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerWizardryBig BrainSolve My Problem

Auto-finds plan pages in 200-sheet bid sets and extracts door counts accurately.

Strengths
  • •Ensemble CV approach handles noisy as-builts better than OCR-only tools.
  • •Generates real-world scale 3D GLB models directly from 2D PDFs.
  • •Pixel-accurate measurement of door and window sizes from scans.
Weaknesses
  • •Requires manual dataset annotation for new model training cycles.
  • •Video demo is unlisted on YouTube, limiting accessibility.
Category
Productivity
Target Audience

Construction estimators, architects, engineers

Similar To

PlanSwift · Bluebeam Revu · Autodesk Takeoff

Post Description

I built QuantTakeoff and releasing v1.0 for validation: Input a construction PDF, get back takeoff report with wall lengths, areas, door/window counts and sizes 3D GLB of the building at real-world scale all under ~10 mins.

The pain it solves: Reduce time estimator to trace out elements either by hand or software and extract out reports.

Stack: Ensemble of computer vision tools (95%), VLM OCR (5%)

Hard parts that are working: Auto-find the plan page in a 200-sheet bid set (most stacks make you point at the right sheet manually). I am still manually annotating a new dataset for the new models. Holds up on noisy as-builts and hand-marked sheets that break OCR-first pipelines Post CV processing to optimize for usability. Pixel accuracy to measure elements including size / width of doors and windows.

Demo: https://youtu.be/fVy7tDFqR98

Particularly want feedback on: Real world estimators feedback on what could be better for the practice. What would make use this or what is missing?

P.S Live demo has been removed to manage compute costs from a few enthusiasts who burn through the HF spaces bill.

Similar Projects