Implementation of Stereo Matching Using High Level Compiler for Parallel Computing Acceleration

Jinglin Zhang (1, 2) , Jean François Nezan (1, 2) , Jean-Gabriel Cousin (1, 2) , Erwan Raffin (3)

1 UEB - Université européenne de Bretagne - European University of Brittany
2 IETR - Institut d'Électronique et des Technologies du numéRique
3 CAPS entreprise

Heterogeneous computing system increases the performance of parallel computing in many domain of general purpose computing with CPU, GPU and other accelerators. With Hardware developments, the software developments like Compute Unified Device Architecture(CUDA) and Open Computing Language (OpenCL) try to offer a simple and visualized tool for parallel computing. But it turn out to be more difficult than programming on CPU platform for optimization of performance. For one kind of parallel computing application, there are different configuration and parameters for various hardware platforms. In this paper, we apply the Hybrid Multi-cores Parallel Programming(HMPP) to automatic-generates tunable code for GPU platform and show the result of implementation of Stereo Matching with detailed comparison with C code version and manual CUDA version. The experimental results show that the default and optimized HMPP have the approximative 1 compared with CUDA implementation. And the HMPP workbench can greatly reduce the time of application development using parallel computing device.

Domaines: Multimédia [cs.MM]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Jinglin Zhang, Jean François Nezan, Jean-Gabriel Cousin, Erwan Raffin. Implementation of Stereo Matching Using High Level Compiler for Parallel Computing Acceleration. 27th Image and Vision Computing New Zealand (IVCNZ), Nov 2012, Dunedin, New Zealand. pp.NC. ⟨hal-00763852⟩