Computer Vision – Object Detection for Construction Blueprints (Training + API Output) - Contract to Hire

Hi, I’m looking for an experienced Computer Vision / Machine Learning engineer who has previously worked with construction blueprints, floor plans, or technical drawings. My goal is to automatically detect and extract detailed information from architectural drawings similar to: Example tasks include detection of: total square footage room segmentation number of rooms doors / windows kitchen & washroom locations furniture symbols (optional) electrical sockets / lighting symbols annotations and dimension text Required Experience Please only apply if you have hands-on experience in at least one of the following: Computer Vision models trained on technical drawings Object detection / instance segmentation Floor-plan / blueprint parsing OCR for annotation and dimension reading Roboflow, YOLOv8, Detectron, Mask-RCNN, SAM or similar CAD / PDF to vector conversion (bonus) Deliverables A working baseline model (trained or fine-tuned) Detection of main objects (rooms, doors, kitchen, etc.) JSON output of detected elements Ability to deploy via API or inference script I am open to: Training a new model Adapting an existing open-source model Using your already-trained blueprint models Paid third-party APIs if appropriate Nice to Have BIM / IFC experience Architectural or engineering background Experience with CubiCasa data, MLStructFP, etc Please include in your proposal Relevant past CV / blueprint projects Which models or datasets you would use If you’ve worked on construction drawings before Estimated time and cost range Important This is not a generic computer-vision task. Must have blueprint / floor-plan or engineering drawing experience. Thanks, Apply tot his job