A NEW SUPER RESOLUTION TECHNIQUE FOR RANGE DATA

Current Time-of-Flight matrix sensors allow for the acquisition of range maps, i.e. surface data, at video rate. This possibility is rather intriguing in light of its applications to 3D video and general motion capture. Unfortunately the resolution of the range maps acquired by such devices is quite limited specially if compared with standard color cameras. This makes highly desirable methods for improving the resolution of range maps. The characteristics of range data are completely different from those of images since geometry and light intensity are physically different quantities, therefore methods suited to image interpolation do not perform well with range data. This work presents an original interpolation technique that exploits side information from a standard color camera to increase the resolution of range maps and demonstrates its performance.

[1]  Sebastian Thrun,et al.  High-quality scanning using time-of-flight depth superresolution , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[2]  Pietro Zanuttigh,et al.  Compression of depth information for 3D rendering , 2009, 2009 3DTV Conference: The True Vision - Capture, Transmission and Display of 3D Video.

[3]  Ahmet M. Kondoz,et al.  Automatic video object segmentation using depth information and an active contour model , 2008, 2008 IEEE 10th Workshop on Multimedia Signal Processing.

[4]  Ballan,et al.  3D Passive shape recovery from texture and silhouette information , 2006 .

[5]  Ruigang Yang,et al.  Spatial-Depth Super Resolution for Range Images , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Ryan Crabb,et al.  Real-time foreground segmentation via range and color imaging , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[7]  Sebastian Thrun,et al.  An Application of Markov Random Fields to Range Sensing , 2005, NIPS.