|
Agency contact Bob Decker Redpines +1 415 409-0233 |
EEMBC contact Peter Torelli, President and CTO EEMBC +1 (203) 423-3179 |
Editor
resources:
Link to
more information:
https://www.eembc.org/adasmark/index.php
ADASMark™
Autonomous Driving Benchmark Suite Is Now Available for Licensing From EEMBC
Benchmarks
Measure Performance of SoCs Used in ADAS Image Processing and Object Detection
PORTLAND, Ore. — July 25, 2018 — EEMBC, an industry consortium that
develops benchmarks for embedded hardware, today announced that its ADASMark™ autonomous
driving benchmark suite is now available for licensing. The benchmarks provide
a performance measurement and optimization tool for automotive companies
building next-generation advanced driver-assistance systems (ADAS).
Intended
to analyze the performance of SoCs used in ADAS implementations above Level 2 —
which require compute-intensive object-detection and visual classification
capabilities — ADASMark uses real-world workloads that represent highly
parallel applications, such as surround view stitching, contour detection, and
convolutional neural-net (CNN) traffic sign classification. The ADASMark benchmarks stress
various forms of compute resources, such as the CPU, GPU, and hardware
accelerators, and thus allow the user to determine the optimal utilization of
available compute resources.
Key features of the ADASMark Benchmark Suite include
an OpenCL 1.2 Embedded Profile API to ensure consistency between compute
implementations; application flows created by a series of micro-benchmarks that
measure and report performance for SoCs handling computer vision, autonomous
driving, and mobile imaging tasks; and a traffic sign recognition CNN inference
engine created by Au-Zone Technologies.
“While more and more automotive embedded systems are
deploying multiple cores, there are still very few frameworks that can utilize
their asymmetric compute resources,” said Peter Torelli, EEMBC president and
CTO. “Representing the next step in the evolution of EEMBC’s automotive
benchmarks and multicore frameworks, ADASMark addresses this issue for the
visual component of ADAS. It uses a new framework with a more relevant workload
while still taking advantage of the GPU hardware in vehicles.”
To request an ADASMark
license, please visit https://www.eembc.org/memberinfo/requestinfo.php?reg=LIC&suite=adasmark.
# # #
About EEMBC
EEMBC
develops performance benchmarks for the hardware and software used in autonomous
driving, mobile imaging, Internet of Things, mobile devices, and many other
applications. Benchmark suites are developed in a consensual process by EEMBC
member companies and EEMBC technical staff to ensure fairness of approach and industry-wide
acceptance. Further information is available at www.eembc.org.
ADASMark
is a trademark and EEMBC is a registered trademark of the Embedded
Microprocessor Benchmark Consortium. All other trademarks appearing herein are
the property of their respective owners.