Abstract | ||
---|---|---|
Real-time stereo matching, which is important in many applications like self-driving cars and 3-D scene reconstruction, requires large computation capability and high memory bandwidth. The most time-consuming part of stereo-matching algorithms is the aggregation of information (i.e. costs) over local image regions. In this paper, we present a generic representation and suitable implementations for three commonly used cost aggregators on many-core processors. We perform typical optimizations on the kernels, which leads to significant performance improvement (up to two orders of magnitude). Finally, we present a performance model for the three aggregators to predict the aggregation speed for a given pair of input images on a given architecture. Experimental results validate our model with an acceptable error margin (an average of 10.4%). We conclude that GPU-like many-cores are excellent platforms for accelerating stereo matching. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/ICPADS.2012.71 | ICPADS |
Keywords | Field | DocType |
3-d scene reconstruction,computation capability,self-driving cars,image matching,stereo matching,cost aggregation,performance optimization,performance modeling,3d scene reconstruction,graphics processing units,generic representation,performance improvement,gpu-like many-cores,aggregation speed,significant performance improvement,computational complexity,performance model,cost aggregators,acceptable error margin,local image regions,accelerating cost aggregation,cost aggregation acceleration,memory bandwidth,gpus,stereo image processing,excellent platform,opencl,real-time stereo matching | Computer science,High memory,Parallel computing,Image processing,Algorithm,Real-time computing,Bandwidth (signal processing),Order of magnitude,Margin of error,Computational complexity theory,Computation,Performance improvement | Conference |
ISSN | ISBN | Citations |
1521-9097 E-ISBN : 978-0-7695-4903-3 | 978-0-7695-4903-3 | 4 |
PageRank | References | Authors |
0.39 | 16 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jianbin Fang | 1 | 265 | 25.31 |
Ana Lucia Varbanescu | 2 | 520 | 44.83 |
Jie Shen | 3 | 101 | 8.05 |
Henk J. Sips | 4 | 1611 | 142.06 |
Gorkem Saygili | 5 | 78 | 6.36 |
van der maaten | 6 | 763 | 48.75 |