ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Depth based movable display using projector Kinect system

Manevarthe, Bhargava and Kalpathi, Ramakrishnan (2018) Depth based movable display using projector Kinect system. In: 12th International Conference on Distributed Smart Cameras (ICDSC), SEP 03-04, 2018, Eindhoven, NETHERLANDS.

[img] PDF
Pro_ Int_Con_Dis_Sma_Cam_2018.pdf - Published Version
Restricted to Registered users only

Download (4MB) | Request a copy
Official URL: https://doi.org/10.1145/3243394.3243691

Abstract

This paper deals with projecting display content on an arbitrary sized planar surface which can move within a larger projection area. These type of systems will give greater control of the display to the users. Since projectors are used in the low illumination environment, depth data will give better accuracy compared to RGB camera. A novel vision-based algorithm is presented to detect the planar surface and track its motion using only depth data. Display content is then pre-warped and projected onto the planar surface. Our detection and tracking approach is then extended for multiple planar surfaces wherein different images/videos are projected on different planar surfaces (from a single projector). We demonstrate that our system can project images and videos onto single or multiple planar surfaces in real time with good accuracy. A sketching application using the proposed tracking and projection system is presented as an aid for users to draw the 3D model.

Item Type: Conference Proceedings
Publisher: ASSOC COMPUTING MACHINERY
Additional Information: 12th International Conference on Distributed Smart Cameras (ICDSC), Eindhoven, NETHERLANDS, SEP 03-04, 2018
Keywords: Movable display; Projector Kinect System; 3D model sketching; Planar surface detection
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 07 Feb 2019 08:59
Last Modified: 07 Feb 2019 08:59
URI: http://eprints.iisc.ac.in/id/eprint/61672

Actions (login required)

View Item View Item