Optimization in DRIVE AGX

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
1

Please provide the following info (check/uncheck the boxes after clicking “+ Create Topic”):
Software Version
DRIVE OS Linux 5.2.0
DRIVE OS Linux 5.2.0 and DriveWorks 3.5
NVIDIA DRIVE™ Software 10.0 (Linux)
NVIDIA DRIVE™ Software 9.0 (Linux)
other DRIVE OS version
other

Target Operating System
Linux
QNX
other

Hardware Platform
NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
NVIDIA DRIVE™ AGX Pegasus DevKit (E3550)
other

SDK Manager Version
1.5.0.7774
other

Host Machine Version
native Ubuntu 18.04
other

I am a beginner using DRIVE (just started last week).
I have a program that uses OpenCV. I have tried the program in three platforms: My Ubuntu 20.04 Host machine, a Jetson AGX Xavier and a DRIVE AGX Xavier

The results in performance are puzzling. The program runs in the platforms in this time

Ubuntu Host : 0m31.726 s
Jetson AGX Xavier: 0m52.205 s
DRIVE AGX Xavier: 1m32.028 s

As you can see, surprisingly, the program is much slower in the DRIVE platform.

For reference, both the Ubuntu and the DRIVE uses OpenCV 3.2.0
It seems all platforms use TBB, but not sure if NEON is activated
The OpenCV version in Jetson is 4.1.1

Can someone help me understand this and what can I do to optimize the run in DRIVE?
Any help would be appreciated in things like:

  1. Optimizing and Profiling C++ programs in DRIVE
  2. NEON and VFPV3 as OpenCV options in DRIVE
  3. Alternatives for Computer vision or Image processing in DRIVE
5

Hi aliaga,
Please refer to the following comment mentioned in the link,
“OpenCV is not officially supported nor is it optimized on the Drive platform. These instructions are provided only to help with porting/prototyping and is not maintained by NVIDIA. It is recommended to utilize NVIDIA APIs moving forward.”

-. To profile, please refer to Nsight Systems : NVIDIA Documentation,
-. For image processing, please refer to NvMedia : DRIVE OS Linux