Augmented Carpentry: The Future of Woodworking with AI and Computer Vision

🛠️ Augmented Carpentry: Bridging Manual Craftsmanship and Digital Precision

In an industry where digital precision often comes at the cost of flexibility and accessibility, a new framework called Augmented Carpentry (AC) is reshaping the narrative. Developed by researchers at EPFL and collaborators, this innovative open-source system brings digital feedback and real-time precision into traditional woodworking—without the need for expensive robotic setups.

📐 What is Augmented Carpentry?

Augmented Carpentry integrates computer vision, custom hardware, and real-time feedback into everyday electric woodworking tools. Instead of relying on CNC machines or complex industrial robotics, AC allows human operators to use retrofitted tools—such as saws and drills—guided by digital overlays shown on touchscreens. The system tracks tool positioning in six degrees of freedom (6DoF) and helps the operator execute precise cuts and holes by visually aligning physical actions with 3D models from a CAD environment.

🔧 Components and Workflow

At the heart of AC is a lightweight C++ engine named the Augmented Carpentry Engine (ACE), optimized for UNIX environments. The system uses a mounted touch screen, a wide-angle fish-eye monocular camera, and a small computing unit (Intel NUC) to track the position of the tool and the timber workpiece. The touch display shows a grayscale camera feed with real-time augmented visual cues guiding the worker.

The key steps in the workflow are:

1. Tool Retrofitting: Tools are fitted with magnetic mounts for sensors and screens. Tool heads are calibrated using 3D models.

2. Timber Mapping: Timber elements are tagged with fiducial markers and scanned, creating a digital map for accurate alignment.

3. Execution: The user follows visual cues on-screen to cut or drill according to the CAD model. Errors and deviations are continuously tracked and logged.

🎯 Precision in Practice

To test the system’s real-world applicability, researchers conducted an extensive fabrication experiment involving 166 joints and 108 drill holes across various types of timber joints. Results were promising:

• Joint positioning errors averaged less than 3 mm.

• Joint face execution achieved sub-millimeter accuracy.

• Drilling orientation errors stayed under 1.2°, with positioning within 4 mm—suitable for most timber fastening applications.

Notably, accuracy decreased slightly for beams longer than 3 meters, indicating a need for improved tracking and object-locking for larger elements. Still, the system consistently supported accurate fabrication for most common workshop sizes.

🧠 Augmented, Not Automated

Unlike rigid automation pipelines, Augmented Carpentry empowers the craftsperson. It provides live feedback without enforcing workflow constraints, allowing experienced workers to adapt their decisions in real time. The UI avoids visual clutter, focusing instead on actionable geometric feedback like angles, depths, and alignment indicators. This design promotes trust and collaboration between human skill and machine precision.

🧩 Broader Implications

This system exemplifies a sustainable and scalable approach to digital fabrication:

• Retrofitting instead of replacing tools makes it accessible for SMEs.

• Open-source architecture encourages widespread adoption.

• Real-time feedback improves build quality and accountability.

Augmented Carpentry doesn’t aim to replace the carpenter—it enhances their capabilities. By making digital fabrication more inclusive and adaptable, AC fosters a new generation of intelligent craftsmanship grounded in both tradition and technology.

📚 References

Settimi, A., Gamerro, J., & Weinand, Y. (2025). Augmented Carpentry: Computer Vision-assisted Framework for Manual Fabrication. Ecole Polytechnique Fédérale de Lausanne. Preprint submitted to Automation in Construction. https://arxiv.org/abs/2503.07473

Benligiray, B., Topal, C., & Akinlar, C. (2019). STag: A stable fiducial marker system. Image and Vision Computing, 89, 158–169. https://doi.org/10.1016/j.imavis.2019.06.007

Fazel, A., & Adel, A. (2024). Enhancing construction accuracy, productivity, and safety with augmented reality for timber fastening. Automation in Construction, 166, 105596. https://doi.org/10.1016/j.autcon.2024.105596