Real time stereo vision using exponential step cost aggregation on GPU
Real time stereo vision using exponential step cost aggregation on GPU
复制标题
DOI：
10.1109/icip.2009.5413693
复制
发表时间：
2009-11
期刊：
影响因子：
--
通讯作者：
Wei Yu;Tsuhan Chen;J. Hoe
中科院分区：
文献类型：
--
作者：
Wei Yu;Tsuhan Chen;J. Hoe
在本文中，我们提出了一种基于图形处理单元（GPU）的实时立体视觉局部成本聚合方法。最近的研究表明，基于精心设计的成本聚合策略的本地方法可以胜过许多全球方法。在这些局部聚合方法中，自适应权窗口在实时约束下生成质量最好的视差图，但比其他局部聚合方法速度慢。提出了一种基于指数阶跃信息传播的快速自适应权值聚合方法。基本思想是在几次迭代中从长距离像素传播信息。我们还讨论了在GPU平台上高效实现的重要技术，其速度比直接实现提高10.5倍。与现有的实时自适应权重方法相比，我们的技术在提高精度的情况下减少了一半以上的计算时间。详细的实验结果表明，在精度-速度权衡空间中，我们的方法在现有的实时或近实时立体算法中是pareto最优的。
In this paper, we propose a local cost aggregation approach for real time stereo vision on a graphics processing unit (GPU). Recent research shows that local approaches based on carefully designed cost aggregation strategies can outperform many global approaches. Among those local aggregation approaches, adaptive-weight window produces the best quality disparity map under real-time constraint, but it is slower than other local approaches. We propose a very fast adaptive-weight aggregation method based on exponential step information propagation. The basic idea is to propagate information from long distance pixels within a few iterations. We also discuss important techniques of efficient implementation on GPU platform, which result in 10.5x speed up than a straightforward implementation. Compared to existing real time adaptive-weight approach, our technique reduces the computation time by more than half at improved accuracy. Detailed experimental results show that our technique is Pareto-optimal among existing real time or near real time stereo algorithms in the accuracy-speed trade-off space.