Physical Guides: An Analysis of 3D Sketching Performance on Physical Objects in Augmented Reality

This work is part of the ARPen track of our Personal Photonics project in which we investigate techniques that allow you to construct 3d models in Augmented Reality by using a pen to sketch around existing objects. We measured and compared the drawing performance when following a path specified either by a visual line or a shape on both physical and virtual objects. Results indicate that users deviate 48% less when drawing on a physical object compared to virtual objects. A concave shape on a physical object further enhances that precision further while a concave shape on a virtual object is among the objects where the most deviations were present.

We present our results in a poster at the ACM CHI 2018 conference in April 2018.


Youtube Video


Augmented Reality (AR) lets users sketch 3D designs directly attached to existing physical objects. These objects provide natural haptic feedback whenever the pen touches them, and, unlike in VR, there is no need to digitize the physical object first. Especially in Personal Fabrication, this lets non-professional designers quickly create simple 3D models that fit existing physical objects. We studied how accurately visual lines and concave/convex surfaces let users draw 3D shapes attached to physical vs. virtual objects in AR. Results show that tracing physical objects is 48% more accurate, but takes longer than tracing virtual objects. Concave physical edges further improve accuracy due to their haptic guidance. Our findings provide initial metrics when designing AR sketching systems.

Contact: Philipp Wacker



Philipp Wacker, Adrian Wagner, Simon Voelker, and Jan Borchers. Physical Guides: An Analysis of 3D Sketching Performance on Physical Objects in Augmented Reality. In Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems, CHI EA '18, pages LBW626:1–LBW626:6, New York, NY, USA, April 2018. ACM. Homepage PDF Document BibTeX Entry.

Created by wacker. Last Modification: Monday 16 of April, 2018 12:04:39 by voelker.

Media Computing Group at RWTH Aachen