Skip to main content
Back to challenges
Challenge

Multifamily Adaptive Reuse Layout OCR & Load Parser

Converting structural office plans into residential floor layouts requires parsing legacy paper blueprints and load-bearing column matrices. Build an OCR and document parsing pipeline using OpenCV and Vision AI to map office floor structural constraints to residential apartment layouts. Achieve 88%+ accuracy in column grid detection.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct an automated CAD and OCR extraction pipeline to analyze structural column grids for office-to-residential conversions.

Delivery guide

How work is evaluated

Evaluation

Evaluates accuracy of column center detection and structural grid boundary parsing.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

Grid Integrity Test

Spans detected must align with reasonable structural column grids (20-40 ft)

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

Column Detection Precision

Precision of structural column location detection • target: 0.92 • range: 0.8-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Ingest scanned structural floor plans using OpenCV image processing filters

  • Detect load-bearing columns and core mechanical shafts using object detection

  • Calculate max permissible unit density based on column spacing and window access

  • Generate standardized JSON geometric layouts for architectural layout planning

Resources and assets

Reference links and supporting material

Dataset notes

Scanned architectural floor plans of commercial office towers marked with structural column locations.

Challenge setup

Evaluates accuracy of column center detection and structural grid boundary parsing.

Challenge input

JSON with image_width_px, image_height_px, detected_contours_count

OpenCV

Performs fast image filtering, edge detection, and contour extraction

Evaluated output

JSON with column_count, average_spacing_ft

Evaluation checks
  • Spans detected must align with reasonable structural column grids (20-40 ft)
  • Precision of structural column location detection • target: 0.92 • range: 0.8-1
Configured evaluation evidence
  • Benchmark: BlueprintParseBench
  • Column Detection Precision target: 0.92
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • RAI · Required
  • MLflow · Optional
  • Hume AI · Optional

Evaluation contract

  • Grid Integrity Test · Weight 1
  • Column Detection Precision · Weight 1

Metrics

  • Column Detection Precision · Target 0.92

Benchmarks

  • BlueprintParseBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with image_width_px, image_height_px, detected_contours_count

    Output: JSON with column_count, average_spacing_ft

    Execution harness: Python

Evaluation tools

  • OpenCV · Vision Computer Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Multifamily Adaptive Reuse Layout OCR & Load Parser