Geometric and texture cue based depth-map estimation for 2D to 3D image conversion

We present a novel framework for estimating depth-map in a single 2D image. The proposed framework employs both vanishing point and superpixels as geometric and texture cues, respectively. The combination of both cues can generate a feasible perceptual depth-map, which will be applied to 2D to 3D image conversion.

[1]  E. A. Hendriks,et al.  Converting 2D to 3D: A Survey , 2005 .

[2]  Yong Ju Jung,et al.  A novel 2D-to-3D conversion technique based on relative height-depth cue , 2009, Electronic Imaging.

[3]  Stephen Gould,et al.  Single image depth estimation from predicted semantic labels , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Daniel P. Huttenlocher,et al.  Efficient Graph-Based Image Segmentation , 2004, International Journal of Computer Vision.

[5]  Manbae Kim,et al.  2D to 3D stereoscopic conversion: depth-map estimation in a 2D single-view image , 2007, SPIE Optical Engineering + Applications.

[6]  V. Cantoni,et al.  Vanishing point detection: representation analysis and new approaches , 2001, Proceedings 11th International Conference on Image Analysis and Processing.

[7]  Jitendra Malik,et al.  Learning a classification model for segmentation , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.