Accelerating feature extraction for patch-based Multi-View Stereo algorithm
Shenzhen Inst. of Adv. Integration Technol., Chinese Acad. of Sci., Shenzhen, China
International Conference on Computer Design and Applications (ICCDA), 2010
@inproceedings{zhang2010accelerating,
title={Accelerating feature extraction for patch-based Multi-View Stereo algorithm},
author={Zhang, H. and Xie, Y. and Heng, P.A.},
booktitle={Computer Design and Applications (ICCDA), 2010 International Conference on},
volume={5},
pages={V5–511},
year={2010},
organization={IEEE}
}
In this paper, we present a novel parallel implementation of HARRIS and DOG detector on GPU for feature extractions of Patch-based Multi-View Stereo (PMVS) algorithm in image sequence. With the Compute Unified Device Architecture(CUDA)-enabled GPU, the acceleration is significant and it obtains a 34 times performance boost comparing to a CPU implementation. We adopt the hardware built-in bilinear interpolation of texture to shorten the time for image resampling while doing image transformations. While lots of time can be saved by using our improved PMVS Algorithm with GPU-based feature extraction, experimental results also show that our implementation can obtain fine details for building accurate object and scene models.
July 27, 2011 by hgpu
Recent source codes
SYCL-BLAS: An implementation of BLAS using the SYCL open standard for acceleration on OpenCL devices
