In-Situ CAD Capture

MobileHCI 2016

Aditya Sankar - University of Washington
Steve Seitz  - University of Washington & Google


Abstract -- We present an interactive system to capture CAD-like 3D models of indoor scenes, on a mobile device. To overcome sensory and computational limitations of the mobile platform, we employ an in situ, semi-automated approach and harness the user's high-level knowledge of the scene to assist the reconstruction and modeling algorithms. The modeling proceeds in two stages: (1) The user captures the 3D shape and dimensions of the room. (2) The user then uses voice commands and an augmented reality sketching interface to insert objects of interest, such as furniture, artwork, doors and windows. Our system recognizes the sketches and add a corresponding 3D model into the scene at the appropriate location. The key contributions of this work are the design of a multi-modal user interface to effectively capture the user's semantic understanding of the scene and the underlying algorithms that process the input to produce useful reconstructions.

Our system captures a CAD-like 3D model of an indoor scene, in a few minutes, on a commodity tablet.

Paper: PDF (6.5 MB)

Citation: Aditya Sankar and Steven Seitz. 2016. "In-Situ CAD Capture."
In Proceedings of the 18th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '16). (ACM DL) (BibTeX)

Contact Author: Aditya Sankar, aditya[at]cs[dot]washington[dot]edu

Acknowledgements: This work was supported in part by the National Science Foundation (IIS-1250793), the University of Washington Animation Research Labs, and Google. We would also like to credit the following Sketchup users {Alons, SpinyMcSpleen3264, Herman Miller, Joseph D., KHITCHZ, Webric (Richard), gui, Steve K., Joey, Ryno, anonymous, Michal S., Tanjoodo, swisscows, Meco Office, Yen Ha, Sedus, Steelcase, lcr2, TinkerToy, sdmitch40, Daweeze} and TurboSquid users {viktor7777, Fworx, AG4T, AlpArt, vandoan, CSI Renderers} for the use of their 3D models in this presentation.