Title: Real-Time Depth-Image-Based Rendering for 3DTV Using OpenCL
Abstract: This paper proposes a real-time (performance of at least 30 fps for full-HD video) Depth-Image-based Rendering (DIBR) approach for stereoscopic 3DTV using OpenCL. Many stereoscopic 3DTV, multi-view, and Free-view-point TV (FTV) technologies have been based on DIBR, aiming at saving bandwidth, and to allow for user adaptation in the client-side. Unlike related work, this paper uses OpenCL for all the DIBR steps, including the re-projection in the virtual views (which is commonly performed using OpenGL, even when implemented in GPGPUs). The use of OpenCL-only can, in some cases, outperform the OpenGL z-testing performance. Two execution models have been implemented (per-line parallel, and per-pixel parallel) and tested against standard video-plus-depth test sequences to show the approach performance.
Publication Year: 2014
Publication Date: 2014-01-01
Language: en
Type: book-chapter
Indexed In: ['crossref']
Access and Citation
Cited By Count: 4
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot