Implementations Impact on Iterative Image Processing for Embedded GPU

Autor: Romera, Thomas, Petreto, Andrea, Lemaitre, Florian, Bouyer, Manuel, Meunier, Quentin, Lacassagne, Lionel
Přispěvatelé: Architecture et Logiciels pour Systèmes Embarqués sur Puce (ALSOC), LIP6, Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS)-Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Lacassagne, Lionel
Jazyk: angličtina
Rok vydání: 2021
Předmět:
[INFO.INFO-AR]Computer Science [cs]/Hardware Architecture [cs.AR]
[INFO.INFO-AR] Computer Science [cs]/Hardware Architecture [cs.AR]
[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing
Image Processing
[INFO.INFO-SE] Computer Science [cs]/Software Engineering [cs.SE]
[INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS]
GPU
[INFO.INFO-DS] Computer Science [cs]/Data Structures and Algorithms [cs.DS]
02 engineering and technology
Energy Consumption
[INFO.INFO-SE]Computer Science [cs]/Software Engineering [cs.SE]
[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM]
[SPI.AUTO]Engineering Sciences [physics]/Automatic
[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
TV-L1
[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing
0202 electrical engineering
electronic engineering
information engineering
[INFO.INFO-DC] Computer Science [cs]/Distributed
Parallel
and Cluster Computing [cs.DC]
[INFO.INFO-RB]Computer Science [cs]/Robotics [cs.RO]
Embedded System
[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing
Optical flow
[INFO.INFO-AO]Computer Science [cs]/Computer Arithmetic
[INFO.INFO-RB] Computer Science [cs]/Robotics [cs.RO]
[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
020202 computer hardware & architecture
[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM]
[SPI.AUTO] Engineering Sciences [physics]/Automatic
[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV]
[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV]
020201 artificial intelligence & image processing
[INFO.INFO-AO] Computer Science [cs]/Computer Arithmetic
[INFO.INFO-DC]Computer Science [cs]/Distributed
Parallel
and Cluster Computing [cs.DC]
[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Zdroj: European Signal Processing Conference (EUSIPCO)
European Signal Processing Conference (EUSIPCO), Aug 2021, Dublin, Ireland
Popis: International audience; The emergence of low-power embedded Graphical Processing Units (GPUs) with high computation capabilities has enabled the integration of image processing chains in a wide variety of embedded systems. Various optimisation techniques are however needed in order to get the most out of an embedded GPU. This paper explores several optimisation methods for iterative stencil-like image processing algorithms on embedded NVIDIA GPUs using the Compute Unified Device Architecture (CUDA) API. We chose to focus our architectural optimisations on the TV-L1 algorithm, an optical flow estimation method based on total variation (TV) regularisation and the L1 norm. It is widely used as a model for more complex optical flow estimations and is used in many recent video processing applications. In this work we evaluate the impact of architecture-oriented optimisations on both execution time and energy consumption on several Nvidia Jetson GPU embedded boards. Results show a speedup up to 3× compared to State-of-the-Art versions as well as a 2.6× decrease in energy consumption.
Databáze: OpenAIRE
Externí odkaz: