Real-time 3D space understanding is becoming prevalent across a wide range of applications and hardware platforms. In order to meet the desired Quality of Service (QoS), computer vision applications tend to be heavily parallelized and exploit any available hardware accelerators. Current approaches in achieving real-time computer vision, evolve around programming languages typically associated with High Performance Computing along with binding extensions for OpenCL or CUDA execution.
Such implementations, although high performing, lack portability across the wide range of diverse hardware resources and accelerators. In this paper, we showcase how a complex computer vision application can be implemented within a managed runtime system. We discuss the complexities of achieving high-performing and portable execution across embedded and desktop configurations. Furthermore, we demonstrate that it is possible to achieve the QoS target of over 30 frames per second (FPS) by exploiting FPGA and GPGPU acceleration transparently through the managed runtime system.
|Presentation Slides (CKotselidis_HeteroMREs_VEE_2017.pdf)||2.74MiB|
Sat 8 AprDisplayed time zone: Azores change
14:00 - 15:30
|Catalyst: GPU-assisted rapid memory deduplication in virtualization environments|
Anshuj Garg Indian Institute of Technology, Bombay, Debadatta Mishra Indian Institute of Technology, Bombay, Purushottam Kulkarni Indian Institute of Technology, BombayFile Attached
|Just-In-Time GPU Compilation for Interpreted Languages with Partial Evaluation|
Juan Fumero The University of Edinburgh, Michel Steuwer The University of Edinburgh, Lukas Stadler Oracle Labs, Austria, Christophe Dubach University of EdinburghLink to publication
|Heterogeneous Managed Runtime Systems: A Computer Vision Case Study|
Christos Kotselidis The University of Manchester, James Clarkson The University of Manchester, Andrey Rodchenko The University of Manchester, Andrew Nisbet The University of Manchester, John Mawer The University of Manchester, Mikel LujánFile Attached