Augmented reality system for rectangular objects

This paper presents a new augmented reality (AR) system for rectangular targets. Unlike conventional methods focusing on text-related objects or a few pre-registered pictures, the proposed method detects rectangular objects in images and provides relevant information with AR features. To this end, we detect and segment rectangle objects with color and edge information, and perform rectification from the extracted boundary. Using the rectified images, relevant information is selected and shown in the original images by considering camera poses. Experimental results show that the proposed AR system is able to provide pleasant user experience and can be adopted in many AR-related applications.

[1]  Anton van den Hengel,et al.  Interactive modelling for AR applications , 2010, 2010 IEEE International Symposium on Mixed and Augmented Reality.

[2]  Veronica Teichrieb,et al.  Texture-less planar object detection and pose estimation using Depth-Assisted Rectification of Contours , 2012, 2012 IEEE International Symposium on Mixed and Augmented Reality (ISMAR).

[3]  Hyung Il Koo.  Segmentation and Rectification of Pictures in the Camera-Captured Images of Printed Documents , 2013, IEEE Transactions on Multimedia.

[4]  Gang Hua,et al.  Automatic Business Card Scanning with a Camera , 2006, 2006 International Conference on Image Processing.

[5]  Subhadip Basu,et al.  Segmentation of Camera Captured Business Card Images for Mobile Devices , 2011, ArXiv.

[6]  Hideo Saito,et al.  Calibration-free projector-camera system for spatial augmented reality on planar surfaces , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[7]  Daijin Kim,et al.  Business card region segmentation by block-based line fitting and largest quadrilateral search with constraints , 2015, 2015 9th International Symposium on Image and Signal Processing and Analysis (ISPA).