Migrating real-time depth image-based rendering from traditional to next-gen GPGPU
Sammy Rogmans,Maarten Dumont,Gauthier Lafruit,Philippe Bekaert +3 more
- 04 May 2009
- pp 1-4
TL;DR: This paper wants to sensitize other researchers to reconsider before migrating their implementation completely, and use the practical migration rules to achieve maximum performance with minimal effort.
read more
Abstract: This paper focuses on the current revolution in using the GPU for general-purpose computations (GPGPU), and how to maximally exploit its powerful resources. Recently, the advent of next-generation GPGPU replaced the traditional way of exploiting the graphics hardware. We have migrated real-time depth image-based rendering - for use in contemporary 3DTV technology - and noticed however that using both GPGPU paradigms leads to a higher performance than non-hybrid implementations. Using this paper, we want to sensitize other researchers to reconsider before migrating their implementation completely, and use our practical migration rules to achieve maximum performance with minimal effort.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Figures
Fig. 4. The application chain of depth image-based rendering takes in 2 input images, and outputs a number of requested intermediate images. 
Fig. 6. Performing occlusion handling in (a) the traditional, and (b) the next-gen GPGPU paradigm. 
Fig. 5. Performing a convolution in (a) the traditional, and (b) the next-gen GPGPU paradigm. 
Fig. 3. The next-generation paradigm exposes the graphics hardware as a generic coprocessor, using a distributed-shared memory model. 
Fig. 2. The traditional paradigm exposes the graphics hardware as a pipeline with four-component vector processors. 
Fig. 7. (a) Individual kernel performance using both traditional and next-gen GPGPU, and (b) the profit by using either paradigm.
Citations
•Proceedings Article
Real-time Video-based View Interpolation of Soccer Events using Depth-selective Plane Sweeping
Patrik Goorts,Cosmin Ancuti,Maarten Dumont,S. Rogmans,Philippe Bekaert +4 more
- 01 Jan 2013
TL;DR: A novel technique to synthesize virtual camera viewpoints for soccer events using a combination of NVIDIA’s shaders language Cg and CUDA to achieve real-time video-based rendering results.
16
Real-time free-viewpoint DIBR on GPUs for 3DTV systems
German Bravo,Luat Do,Svitlana Zinger +2 more
- 29 Sep 2011
TL;DR: Using a combination of the highly parallel programming architecture CUDA and a graphics API, this work has achieved a real-time performance operating on 1080p HD multi-view video with a rendering quality that is comparable to the software implementation.
10
Real-time free-viewpoint DIBR on GPUs for large base-line multi-view 3DTV videos
Luat Do,German Bravo,Svitlana Zinger +2 more
- 29 Dec 2011
TL;DR: This paper reports on the implementation of an efficient free-viewpoint DIBR algorithm with an off-the-shelf GPU, which can be readily integrated into future 3DTV systems with a real-time performance operating on 1080p HD multi-view video with a rendering quality that is comparable to the software implementation.
10
Biological-aware stereoscopic rendering in free viewpoint technology using GPU computing
Sammy Rogmans,Maarten Dumont,Gauthier Lafruit,Philippe Bekaert +3 more
- 07 Jun 2010
TL;DR: A biological-aware stereoscopic renderer that is used in a video communication system, to convincingly provide the participants with synthetic 3D perception that is experienced as being rich and natural, without loosing any visual comfort whatsoever.
Intelligent Visual Supercomputing on Hybrid Graphical Multiprocessor Environments
S. Rogmans
- 01 Jan 2013
TL;DR: In this paper, the authors investigate the principles of efficiently utilizing massively available parallel computational power, i.e. more specifically the resources that are available in the Graphics Processing Unit (GPU), and propose a set of transformation rules to indicate the use of different computational models depending on the size of the convolution kernel.
3
References
A taxonomy and evaluation of dense two-frame stereo correspondence algorithms
TL;DR: This paper has designed a stand-alone, flexible C++ implementation that enables the evaluation of individual components and that can easily be extended to include new algorithms.
•Proceedings Article
A Survey of General-Purpose Computation on Graphics Hardware.
John D. Owens,David Luebke,Naga K. Govindaraju,Mark J. Harris,Jens Krüger,Aaron Lefohn,Timothy John Purcell +6 more
- 01 Jan 2005
TL;DR: The techniques used in mapping general-purpose computation to graphics hardware will be generally useful for researchers who plan to develop the next generation of GPGPU algorithms and techniques.
2.1K
Optimization principles and application performance evaluation of a multithreaded GPU using CUDA
Shane Ryoo,Christopher I. Rodrigues,Sara S. Baghsorkhi,Sam S. Stone,David B. Kirk,Wen-mei W. Hwu +5 more
- 20 Feb 2008
TL;DR: This work discusses the GeForce 8800 GTX processor's organization, features, and generalized optimization strategies, and achieves increased performance by reordering accesses to off-chip memory to combine requests to the same or contiguous memory locations and apply classical optimizations to reduce the number of executed operations.
Image Convolution with CUDA
Victor Podlozhnyuk
- 01 Jan 2007
TL;DR: This document shows how a separable convolution filter can be implemented in NVIDIA CUDA and provides some guidelines for performance optimizations.
•Proceedings Article
Efficient stream reduction on the GPU
David Roger,Ulf Assarsson,Nicolas Holzschuch +2 more
- 04 Oct 2007
TL;DR: This paper presents a new efficient algorithm for stream reduction on the GPU that works by splitting the input stream into smaller components of a fixed size, on which it run a standard stream reduction pass with line drawing.