An Open Benchmark Implementation for Multi-CPU Multi-GPU Pedestrian Detection in Automotive Systems
URL: http://ieeexplore.ieee.org/document/8203793/
Authors: Trompouki, Matina / Kosmidis, Leonidas / Navarro, Nacho
Research Lines: Application optimization for GPU acceleration / COTS multicore real-time systems / Data Placement for Heterogeneous Memory Systems / Message Passing Interface (MPI)
Publication: 2017 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)
Place Published: Irvine, California
Pagination: 305-312
Paraules clau: ADAS software, Advanced Driving Assistance Systems, Automotive engineering, automotive industry, automotive processors, automotive systems, Benchmark testing, benchmarking candidate platforms, computational power, Data structures, driver information systems, Feature extraction, GPU implementation, Graphics processing units, image recognition, Industries, multi-CPU multi-GPU pedestrian detection, multicore CPUs, multiple GPUs, multiprocessing systems, multiprocessor system, Nvidia GPUs, Object detection, open benchmark implementation, open implementation, parallel architectures, pedestrian detection benchmark, pedestrians, programming models, Resource management, traffic engineering computing, Viola-Jones image recognition algorithm