Real-Time 3D Depth Generation for Stereoscopic Video Applications with Thread-Level Superscalar-Pipeline Parallelization

Authors:
Guo-An Jian;Cheng-An Chien;Peng-Sheng Chen;Jiun-In Guo
Affiliations:
Department of Computer Science and Information Engineering, National Chung Cheng University, Chia-Yi, Taiwan;Department of Computer Science and Information Engineering, National Chung Cheng University, Chia-Yi, Taiwan;Department of Computer Science and Information Engineering, National Chung Cheng University, Chia-Yi, Taiwan;Department of Electronics Engineering, National Chiao Tung University, HsinChu, Taiwan
Venue:
Journal of Signal Processing Systems
Year:
2013

Citing 10
Cited 0

Advanced compiler optimizations for supercomputers

Communications of the ACM - Special issue on parallelism
Use of the Hough transformation to detect lines and curves in pictures

Communications of the ACM
Mean Shift Analysis and Applications

ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Digital photography with flash and no-flash image pairs

ACM SIGGRAPH 2004 Papers
3D Stereoscopic Image Pairs by Depth-Map Generation

3DPVT '04 Proceedings of the 3D Data Processing, Visualization, and Transmission, 2nd International Symposium
Computer Architecture, Fourth Edition: A Quantitative Approach

Computer Architecture, Fourth Edition: A Quantitative Approach
Patterns for parallel programming

Patterns for parallel programming
Creating Depth Map from 2D Scene Classification

ICICIC '08 Proceedings of the 2008 3rd International Conference on Innovative Computing Information and Control
Make3D: Learning 3D Scene Structure from a Single Still Image

IEEE Transactions on Pattern Analysis and Machine Intelligence
A novel 2Dd-to-3D conversion system using edge information

IEEE Transactions on Consumer Electronics

Quantified Score

Hi-index	0.00

Visualization

Abstract

We propose a low-complexity algorithm for stereoscopic video applications that generates a high-quality 3D image depth map from a single 2D image. Based on their characteristics, 2D images are classified into one of three categories before being processed by the proposed low-complexity algorithm to generate corresponding depth maps. We also extend the 3D depth algorithm to construct a parallel 3D video system. A thread-level superscalar-pipelining approach is developed to parallelize the 3D video system. Experimental results for HD1080 resolution images demonstrate that the algorithm can generate high-quality depth maps with an average reduction in the computational complexity of 98.2 % compared with a conventional algorithm. The parallel 3D video system can achieve a processing speed of 63.66 fps for HD720 resolution video.