26 June 2017 Depth image super-resolution via semi self-taught learning framework
Author Affiliations +
Abstract
Depth images have recently attracted much attention in computer vision and high-quality 3D content for 3DTV and 3D movies. In this paper, we present a new semi self-taught learning application framework for enhancing resolution of depth maps without making use of ancillary color images data at the target resolution, or multiple aligned depth maps. Our framework consists of cascade random forests reaching from coarse to fine results. We learn the surface information and structure transformations both from a small high-quality depth exemplars and the input depth map itself across different scales. Considering that edge plays an important role in depth map quality, we optimize an effective regularized objective that calculates on output image space and input edge space in random forests. Experiments show the effectiveness and superiority of our method against other techniques with or without applying aligned RGB information
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Furong Zhao, Zhiguo Cao, Yang Xiao, Xiaodi Zhang, Ke Xian, Ruibo Li, "Depth image super-resolution via semi self-taught learning framework", Proc. SPIE 10332, Videometrics, Range Imaging, and Applications XIV, 103320R (26 June 2017); doi: 10.1117/12.2270079; https://doi.org/10.1117/12.2270079
PROCEEDINGS
11 PAGES


SHARE
RELATED CONTENT

Super-resolution PMD camera for applied metrology
Proceedings of SPIE (March 03 2016)
Trinocular vision: a 3-D solution
Proceedings of SPIE (March 01 1992)
Three-dimensional gauging with stereo computer vision
Proceedings of SPIE (September 01 1991)
Point matching of two images under large different resolution
Proceedings of SPIE (November 15 2007)
High-performance camera calibration algorithm
Proceedings of SPIE (October 22 1993)

Back to Top