Benchmarking the NXP i.MX8M+ neural processing unit: smart parking case study
Article Sidebar
Main Article Content
Abstract
Nowadays, deep learning has become one of the most popular solutions for computer vision, and it has also included the Edge. It has influenced the System-on-Chip (SoC) vendors to integrate accelerators for inference tasks into their SoCs, including NVIDIA, NXP, and Texas Instruments embedded systems. This work explores the performance of the NXP i.MX8M Plus Neural Processing Unit (NPU) as one of the solutions for inference tasks. For measuring the performance, we propose an experiment that uses a GStreamer pipeline for inferring license plates, which is composed of two stages: license plate detection and character inference. The benchmark takes execution time and CPU usage samples within the metrics when running the inference serially and parallel. The results show that the key benefit of using the NPU is the CPU freeing for other tasks. After offloading the license plate detection to NPU, we lowered the overall CPU consumption by 10x. The performance obtained has an inference rate of 1 Hz, limited by the character inference.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Los autores conservan los derechos de autor y ceden a la revista el derecho de la primera publicación y pueda editarlo, reproducirlo, distribuirlo, exhibirlo y comunicarlo en el país y en el extranjero mediante medios impresos y electrónicos. Asimismo, asumen el compromiso sobre cualquier litigio o reclamación relacionada con derechos de propiedad intelectual, exonerando de responsabilidad a la Editorial Tecnológica de Costa Rica. Además, se establece que los autores pueden realizar otros acuerdos contractuales independientes y adicionales para la distribución no exclusiva de la versión del artículo publicado en esta revista (p. ej., incluirlo en un repositorio institucional o publicarlo en un libro) siempre que indiquen claramente que el trabajo se publicó por primera vez en esta revista.
References
"i.MX8M Plus”, Nxp.com, 2021. [Online]. Available: https://www.nxp.com/products/processors-and-microcontrollers/arm-processors/i-mx-applications-processors/i-mx-8-processors/i-mx-8m-plus-arm-cortex-a53-machine-learning-vision-multimedia-and-industrial-iot:IMX8MPLUS. [Accessed: 22- Sep- 2021].
NVIDIA. 2014. DATA SHEET NVIDIA Jetson Nano System-on-Module. [Online]. Available: https://developer. nvidia.com/embedded/jetson-nano [Accesed: 22- Mar- 2022]
H. Bouzidi, H. Ouarnoughi, S. Niar, and S. Ait El Cadi. 2022. Performances Modeling of Computer Vision-based CNN on Edge GPUs. ACM Trans. Embed. Comput. Syst. DOI: https://doi.org/10.1145/3527169
“GstInference Benchmarks”, RidgeRun Embedded Solutions LLC. [Online]. Available: https://developer.ridgerun.com/wiki/index.php?title=GstInference/Benchmarks. [Accessed: 14- Sep- 2021].
“Introduction to GstInference”, developer.ridgerun.com. [Online]. Available: https://developer.ridgerun.com/ wiki/index.php?title=GstInference/Introduction. [Accessed: 20- Oct- 2021].
“Neural Networks API | Android NDK | Android Developers”, Android Developers. [Online]. Available: https:// developer.android.com/ndk/guides/neuralnetworks. [Accessed: 09- Sep- 2021].
V. Sivakumar, A. Gordo and M. Paluri, “Rosetta: Understanding text in images and videos with machine learning”, Facebook Engineering, 2021. [Online]. Available: https://engineering.fb.com/2018/09/11/ai-research/ rosetta-understanding-text-in-images-and-videos-with-machine-learning/. [Accessed: 20- Sep- 2021].
“GStreamer: open source multimedia framework”, Gstreamer.freedesktop.org. [Online]. Available: https:// gstreamer.freedesktop.org/. [Accessed: 14- Sep- 2021].